The Arabidopsis MIK2 receptor elicits immunity by sensing a conserved signature from phytocytokines and microbes

Sessile plants encode a large number of small peptides and cell surface-resident receptor kinases, most of which have unknown functions. Here, we report that the Arabidopsis receptor kinase MALE DISCOVERER 1-INTERACTING RECEPTOR-LIKE KINASE 2 (MIK2) recognizes the conserved signature motif of SERINE-RICH ENDOGENOUS PEPTIDEs (SCOOPs) from Brassicaceae plants as well as proteins present in fungal Fusarium spp. and bacterial Comamonadaceae, and elicits various immune responses. SCOOP signature peptides trigger immune responses and altered root development in a MIK2-dependent manner with a sub-nanomolar sensitivity. SCOOP12 directly binds to the extracellular leucine-rich repeat domain of MIK2 in vivo and in vitro, indicating that MIK2 is the receptor of SCOOP peptides. Perception of SCOOP peptides induces the association of MIK2 and the coreceptors SOMATIC EMBRYOGENESIS RECEPTOR KINASE 3 (SERK3) and SERK4 and relays the signaling through the cytosolic receptor-like kinases BOTRYTIS-INDUCED KINASE 1 (BIK1) and AVRPPHB SUSCEPTIBLE1 (PBS1)-LIKE 1 (PBL1). Our study identifies a plant receptor that bears a dual role in sensing the conserved peptide motif from phytocytokines and microbial proteins via a convergent signaling relay to ensure a robust immune response.


Results
The kinase domain of MIK2 elicits specific responses. We have previously identified RLK7 as the receptor of the secreted peptides PIP1 and PIP2 19 . RLK7 belongs to the subfamily XI of LRR-RKs, including PEPRs and MIK2 (Supplementary Fig 1a). Similar to RLK7 and PEPR1, MIK2 is transcriptionally upregulated upon treatments of flg22, a synthetic 22-amino-acid peptide derived from bacterial flagellin, and elf18, a synthetic 18-amino-acid peptide derived from bacterial EF-Tu ( Supplementary Fig. 1a). To explore the specific function of the extracellular and the cytosolic kinase domains of LRR-RKs, we generated a chimeric receptor consisting of the extracellular domain (ECD) of RLK7 and the transmembrane and cytoplasmic kinase domains (TK) of MIK2 (RLK7 ECD -MIK2 TK ). The chimeric RLK7 ECD -MIK2 TK and the full-length RLK7 (RLK7 ECD -RLK7 TK ) under the control of the cauliflower mosaic virus (CaMV) 35S promoter were transformed into the rlk7 mutant (Fig. 1a).
Compared to the rlk7 mutant, both RLK7/rlk7 and RLK7 ECD -MIK2 TK /rlk7 transgenic seedlings restored the sensitivity to PIP1 treatment exhibiting root growth retardation (Fig. 1b, c), indicating that the chimeric RLK7 ECD -MIK2 TK receptor is functional. Notably, RLK7 ECD -MIK2 TK /rlk7 seedlings showed more substantial root growth retardation to PIP1 treatment than RLK7/rlk7 (Fig. 1b, c). Besides, PIP1 treatment caused brown roots and darkened root-hypocotyl junctions in RLK7 ECD -MIK2 TK /rlk7 seedlings, but not in RLK7/rlk7 seedlings (Supplementary Fig. 1b). Furthermore, PIP1 treatment triggered a robust production of ROS in RLK7 ECD -MIK2 TK /rlk7 seedlings compared to that in RLK7/rlk7 seedlings, which weakly induced ROS production ( Fig. 1d) 19 . We also generated a chimeric receptor carrying the ectodomain of PEPR1 and the transmembrane and cytoplasmic kinase domains of MIK2 (PEPR1 ECD -MIK2 TK ) and transformed it into the pepr1/pepr2 (pepr1,2) mutant (Supplementary Fig. 1c). The PEPR1 ECD -MIK2 TK /pepr1,2 transgenic seedlings also showed severe root growth retardation, brown roots, and darkened root-hypocotyl junctions compared to WT seedlings upon Pep1 treatment ( Supplementary Fig. 1d, e). Thus, the MIK2 kinase domain likely triggers specific responses differently from those mediated by the RLK7 and PEPR1 kinase domains. However, we cannot completely rule out the possibility that this was attributed to the increased expression of MIK2 TK in RLK7 ECD -MIK2 TK /rlk7 and PEPR1 ECD -MIK2 TK /pepr1,2 transgenic plants.
MIK2 TK -specific responses prompted us to investigate if MIK2, like RLK7 and PEPR1, acts as a receptor to perceive an endogenous peptide ligand(s). DAMP peptide signaling often induces the expression of themselves or precursors, thereby providing positive feedback for immune signal amplification 10,13 . We thus conducted an RNA-sequencing (RNA-Seq) analysis and focused on small peptide genes. Among 51 small peptide genes induced by PIP1 treatment in RLK7 ECD -MIK2 TK /rlk7, 18 genes encode SCOOPs 23 , SECRETED TRANSMEMBRANE PEPTIDEs (STMPs) 33 , and SCOOP/STMP-like proteins (Fig. 1e, Supplemental Fig. 2a, Supplementary Data 1). SCOOPs and STMPs have been recently identified as two groups of secreted peptides with roles in plant growth and defense 23,33 . SCOOPs and STMPs share certain degrees of similarities at the amino (N)-terminal signal peptide and/or the carboxyl (C)-terminal domain with overlapping family members, such as SCOOP4/STMP10, SCOOP13/ STMP1, and SCOOP14/STMP2 (Supplementary Fig. 2b).
Multiple SCOOP peptides activate diverse plant immune and physiological responses. Fourteen SCOOPs (SCOOP1- 14) have been identified in the Arabidopsis genome 23 . Some peptideencoding genes induced in RLK7 ECD -MIK2 TK /rlk7 seedlings upon PIP1 treatment, such as AT1G36640, AT1G36622, STMP4, 5, and 7, bear sequence similarities with SCOOPs (Fig. 1e). We thus searched the Arabidopsis genome and identified additional nine SCOOP family members, namely SCOOP15-23 (Fig. 2a, Supplementary Fig. 2b). SCOOP15 was also named STMP8 33 . Thus, the Arabidopsis genome encodes at least 23 SCOOP isoforms characterized by a conserved C-terminus with an SxS motif (where S is serine and x is any amino acid) (Fig. 2a, Supplementary  Fig. 2b) 23 . Some SCOOPs, including SCOOP6, 7, 10, 11, and 15, contain two copies (an uppercase A or B was added for each copy) of the conserved domain with an SxS motif (Fig. 2a, Supplementary Fig. 2b). The expression pattern of SCOOP genes differs in various plant tissues (Fig. 2b). Of those, SCOOP10 has the highest expression in mature rosette leaves and shoots of seedlings. In contrast, SCOOP12 and 13 have the highest expression in roots and flowers, respectively (Fig. 2b), suggesting that different SCOOPs may bear tissue-specific functions.
SCOOPs possess an N-terminal signal peptide directing peptide secretion, variable regions, and a C-terminal conserved region ( Supplementary Fig. 2b). Protein structure prediction indicates that SCOOP12 without the signal peptide forms an αhelix followed by a loop-like structure ( Supplementary Fig. 3a). Secreted peptides are typically translated as precursor proteins and subsequently secreted into apoplasts for proteolytic processing 34 . To determine whether SCOOP12 is secreted and processed, we generated pSCOOP12::SCOOP12-GFP/scoop12 transgenic plants, in which SCOOP12 fused with GFP at the C-terminus under the control of its native promoter was transformed into the scoop12 mutant. SCOOP12-GFP signals were detected in the plasma membrane as well as apoplasts in leaf cells of transgenic plants ( Supplementary Fig. 3b). Because the active SCOOP12 peptide is predicted to localize on SCOOP12  ) and MIK2 transmembrane and cytoplasmic kinase domains (MIK2 TK ) and the full-length RLK7 (RLK7 ECD -RLK7 TK ) driven by the CaMV35S promoter were transformed into rlk7. b, c RLK7 ECD -MIK2 TK /rlk7 transgenic seedlings show more severe root growth inhibition to PIP1 treatment than RLK7/rlk7 seedlings. Images of 10-day-old seedlings of Arabidopsis WT (Col-0), rlk7, two representative lines (L9 and L16) of RLK7 ECD -MIK2 TK /rlk7, and two lines (L4 and L5) of RLK7/rlk7 grown on ½MS plates with or without 1 µM PIP1: red stars indicate the root tips, scale bar, 4 mm (b; quantification of root length of seedlings (c)). d PIP1 treatment induces a more pronounced ROS production in RLK7 ECD -MIK2 TK /rlk7 than that in RLK7/rlk7 transgenic seedlings. One-week-old seedlings grown on ½MS plates were treated with H 2 O or 100 nM PIP1, and ROS production was calculated as the total relative luminescence units (RLU) (see "Methods" for details). e PIP1 treatment induces the expression of SCOOP genes in RLK7 ECD -MIK2 TK /rlk7 transgenic plants but not in WT plants. Ten-day-old seedlings grown on ½MS plates were treated with 1 µM PIP1 for 0, 1, and 6 h for RNA-Seq analysis. Heatmap represents adjusted log 2 -transformed transcript levels of SCOOPs and STMPs (see "Methods" for details). The box plots in (c) and (d) show the first and third quartiles as bounds of box, split by the medians (lines), with whiskers extending 1.5-fold interquartile range beyond the box, and minima and maxima as error bar. Different letters indicate a significant difference with others (P < 0.05, one-way ANOVA followed by Tukey's test, n = 28, 26,15,15,15,13,15,16,15,14,15,15 for plots from left to right in (c), n = 3, 4, 3, 4, 3, 5, 3, 5 for plots from left to right in (d)). The experiments in (b-d) were repeated three times with similar results.
C-terminus and separates it from two flanking variable regions ( Supplementary Fig. 3c), we speculate that SCOOP12 might be proteolytically processed at these sites 34 . In line with our prediction, multiple bands were detected for SCOOP12-GFP in transgenic plants by an immunoblotting analysis ( Supplementary  Fig. 3d). Together, our data hint that SCOOP12-GFP is likely secreted and proteolytically processed in Arabidopsis.
To determine the SCOOP12 processing sites, we followed an established protocol 35,36 and incubated the recombinant HIS-SCOOP12 proteins with the protein extracts isolated from Arabidopsis seedlings. Apparently, the HIS-SCOOP12 proteins were proteolytically processed upon incubation with Arabidopsis protein extracts ( Supplementary Fig. 3e). The processed products were subjected to a nano-flow liquid chromatography-tandem mass spectrometry (Nano-LC-MS/MS) analysis. A SCOOP12 fragment from leucine 44 (L 44 ) to lysine 78 (Y 78 ) was identified ( Supplementary Fig. 3f), suggesting that SCOOP12 was cleaved after arginine 43 (R 43 ) in a dibasic cleavage site ( Supplementary  Fig. 3g). Importantly, when the dibasic residues of R 42 R 43 were substituted with alanine residues (A 42 A 43 ), one of the processed bands of SCOOP12-GFP expressed in protoplasts was abolished ( Supplementary Fig. 3h), further implicating the cleavage of   10, and 11 are respectively indicated with A and B. The conserved and similar residues are in red and yellow, respectively. b Expression of SCOOPs in different tissues. Heatmap represents adjusted log 2 transformed transcript levels in 4-week-old leaves (L), 7-week-old flowers (F), and 10-day-old shoots (S) and roots (R). Data represent averages of relative expression levels of SCOOPs normalized to UBQ10 from three independent repeats. c, d SCOOPs inhibit root growth. Images of 10-day-old seedlings with 1 µM peptides: Red stars indicate the root tips, scale bar, 4 mm (c); quantification of root length (d). e SCOOPs cause root browning. Microscopic images of root tips with arrowheads indicating browning. Scale bar, 500 µm. f SCOOPs distort root meristems. Confocal images of root tips of 5-day-old seedlings with 1 µM peptides for 48 h and propidium iodide (PI) staining. Arrowheads indicate boundaries of root meristems and elongation zones. Scale bar, 100 µm. g SCOOPs trigger cytosolic Ca 2+ increases. p35S::Aequorin seedlings were treated with 100 nM SCOOPs. Cytosolic Ca 2+ concentration is indicated as means ± SE of RLU (n = 4, SCOOP10 B and SCOOP12; n = 3, H 2 O, SCOOP13, and SCOOP14). h SCOOPs induce ROS production. One-week-old seedlings were treated with 100 nM peptides, and ROS production was calculated as total RLU. i SCOOPs induce MAPK activation. Ten-day-old seedlings were treated with 1 µM peptides, and MAPK activation was analyzed by α-pERK1/2 immunoblotting (IB). Coomassie brilliant blue (CBB) staining of Rubisco (RBC) for protein loading. Molecular weight (kDa) was labeled on the left of all immunoblots. j SCOOPs induce ethylene production. Ethylene concentration in leaf disks was indicated as parts per million (ppm). k SCOOP12 induces a robust ROS production in roots. Detached roots from 1-week-old seedlings were treated with 100 nM peptides. l SCOOPs induce expression of pMYB51::GUS in roots. Microscopic images of root tips of 1-week-old pMYB51::GUS seedlings treated with 1 µM SCOOPs for 3 h and subjected to GUS staining. Scale bar, 1 mm. Box plots in (d), (h), (j), and (k) show the first and third quartiles as bounds of box, split by the medians (lines), with whiskers extending 1.5-fold interquartile range beyond the box, and minima and maxima as error bar. Different letters indicate a significant difference with others (P < 0.05, one-way ANOVA followed by Tukey's test, n = 10, 11, 11, 10, 11, 11, 11,10, 11, 10, 10, 9 for plots from left to right in (d); n = 3 for plots in (h) and (j); n = 4 for plots in (k). Experiments in (b-e), and (g-l) were repeated three times, and (f) twice with similar results.
SCOOP12 after R 43 . The data are consistent with the predicted cleavage site between the variable region I and the conserved domain ( Supplementary Fig. 3c). We noted that the predicted cleavage site derived from the in vitro assay using plant extracts might not be fully representative of the situation in vivo.
SCOOP12 has been shown to regulate defense response and root elongation in Arabidopsis 23 . We tested whether other SCOOPs also have the activity to induce immune-related responses in Arabidopsis. Based on the sequence characteristics and expression patterns, we synthesized ten SCOOP peptides (SCOOP4, 6 B , 8, 10 B , 12, 13, 14, 15 B , 20, and 23) corresponding to the SxS motif. We also synthesized the STMP4 peptide, which has an amino acid deletion in the conserved SxS motif. Treatment of all synthesized SCOOP peptides, but not STMP4, inhibited root growth to various degrees, with SCOOP10 B and 13 being the most active (Fig. 2c, d). A close-up view indicated that prolonged treatments of SCOOP10 B , 12, 13, and 14 peptides in WT seedlings led to brown roots, especially at the root tips of seedlings (Fig. 2e), and distorted root meristems (Fig. 2f), which were not reported for flg22 or Pep1 peptides 16,37 . The browning sites are different in 35S::RLK7 ECD -MIK2 TK transgenic plants treated with PIP1 ( Supplementary Fig. 1b) and WT plants treated with SCOOPs (Fig. 2e). It is not clear the cause of root browning but could be associated with ROS production and cell wall modifications, such as lignification and suberification, upon SCOOP perception.
Similar to flg22, SCOOP peptides induced various hallmarks of PTI responses, including the cytosolic calcium increase (Fig. 2g), ROS production ( Fig. 2h), MAPK activation (Fig. 2i), and ethylene production (Fig. 2j). SCOOP10 B and 12 exhibited more potent activities than others in triggering PTI responses ( Fig. 2g-i). SCOOP13 and 14 had relatively weak activities for Ca 2+ and ROS bursts but were potent in inhibiting root growth (Fig. 2c, d, g, h). Besides, SCOOP10 B , 12, 13, and 14 triggered the MAPK activation in both shoots and roots ( Supplementary Fig. 3i). SCOOP12induced ROS production in roots was significantly higher than that elicited by flg22 (Fig. 2k). Further, SCOOP10 B , 12, 13, and 14 treatments induced the promoter activities of MYB51, a marker gene for root immunity 38 , in the roots of pMYB51::GUS transgenic plants (Fig. 2l). The data indicate that SCOOPs play a role in immune activation in both shoots and roots.
MIK2 mediates SCOOP-induced responses. Next, we tested if MIK2 is required for SCOOP-triggered responses. We isolated two alleles of T-DNA insertional mutants, mik2-1 (SALK_061769) and mik2-2 (SALK_046987). Both mik2-1 and mik2-2 were insensitive to the treatment of SCOOP10 B (Fig. 3a, b). The root growth retardation became more pronounced with the increased concentrations of SCOOP10 B from 1 to 100 nM, which was completely abolished in mik2-1 (Fig. 3c, d). In addition, the mik2-1 and mik2-2 mutants were insensitive to the growth inhibition triggered by the other nine SCOOPs (Fig. 3e, Supplementary  Fig. 4a). However, the mik2-1 mutant responded to the Pep1 treatment similarly to WT plants ( Supplementary Fig. 4b).
Besides, MIK2-LIKE (MIK2L), the closest homolog of MIK2, and other members of subfamily XI LRR-RKs, including RLK7 19 , PEPR1 17 , PEPR2 18 , HAESA (HAE), and HAESA-LIKE2 (HSL2) 39 , were not involved in responsiveness to SCOOP10 B (Supplementary Fig. 4c). The fls2 mutant responded to SCOOP10 B similarly to WT plants ( Supplementary Fig. 4c), supporting that the activities in synthesized SCOOP peptides were unlikely due to the contamination of flg22. The data indicate that the root growth inhibition triggered by SCOOP peptides genetically depends on MIK2, but not other members of the subfamily XI LRR-RKs. Similar to SCOOPs, the apparent MIK2 orthologs with ≥60% of amino acid identity to Arabidopsis MIK2 were identified only in Brassicaceae plants ( Supplementary Fig. 4d) 23,33 . In contrast to the tissue-specific expression of SCOOPs, MIK2 is highly expressed in shoots, roots, and leaves ( Supplementary Fig. 4e), consistent with a previous report 28 .
We also determined whether SCOOP-mediated other physiological and immune responses depended on MIK2. The SCOOP10 B -and 12-induced ROS production (Fig. 3f) and MAPK activation (Fig. 3g) were abolished in mik2 mutants. The SCOOP10 B -and 12-induced distortion of root meristems was not observed in mik2-1 (Fig. 3h). RNA-Seq analyses indicate that SCOOP12 treatment for 1 h regulates the expression of 2249 genes (1642 upregulated and 607 downregulated, fold change ≥2 or ≤0.5, p-value ≤0.01) in WT plants compared to 24 regulated genes in mik2, suggesting that SCOOP12-regulated gene transcription is almost completely abolished in mik2 ( Supplementary Fig. 4f, Supplementary Data 2). Reverse transcription-quantitative polymerase chain reactions (RT-qPCR) confirmed that the upregulation of three immune-related marker genes, WRKY30, WRKY33, and CYP81F2, by SCOOP10 B , 12, and 13 in WT seedlings was abolished in mik2 mutants (Fig. 3i). Taken together, these data indicate that SCOOP-triggered responses depend on MIK2.
MIK2 is the receptor of SCOOP12. Since SCOOPs trigger MIK2-dependent responses, we investigated whether SCOOP12 could bind to MIK2. We synthesized the red fluorescent tetramethylrhodamine (TAMRA)-labeled SCOOP12 peptides at its N-terminus (TAMRA-SCOOP12) and determined the ability of WT and mik2 plants for binding to TAMRA-SCOOP12 in vivo. TAMRA-SCOOP12 peptides were bioactive as they triggered MIK2-dependent root growth inhibition and MAPK activation, despite to a slightly lesser extent than SCOOP12 ( Supplementary  Fig. 5a-c). Red fluorescent signals were detected in root tips and leaf protoplasts in WT plants, but not in mik2-1 upon treatment of TAMRA-SCOOP12 (Fig. 4a). Importantly, pretreatment of unlabeled SCOOP12, but not flg22, markedly reduced red fluorescent signals of TAMRA-SCOOP12 in WT seedlings (Fig. 4b), indicating specific and MIK2-dependent binding of SCOOP12 in vivo.
To test whether SCOOP12 directly binds to MIK2 in vitro, we employed isothermal titration calorimetry (ITC) analysis using the extracellular LRR domain of MIK2 (MIK2 ECD ) expressed in insect cells. ITC quantitatively measures the binding equilibrium by determining the thermodynamic properties of protein-protein interaction in solution. As shown in Fig. 4c, MIK2 ECD bound SCOOP12 potently with a dissociation constant (K d ) of 3.18 µM. Alanine (A) substitutions on two conserved serine (S) residues in the SxS motif of SCOOP12 (SCOOP12 SS/AA ) abolished its activities to inhibit root growth through MIK2 ( Supplementary  Fig. 5d, e) 23 . No binding was detected between MIK2 ECD and SCOOP12 SS/AA (Fig. 4d). The data indicate that MIK2 recognizes and binds to SCOOP12 directly, and the conserved SxS motif is essential for SCOOP12 binding to MIK2. In addition, we performed surface plasmon resonance (SPR) assays, in which affinity-purified MIK2 ECD proteins isolated from insect cells were immobilized by an amine-coupling reaction on a sensor chip, and synthesized SCOOP12 peptides were used as the flow-through analytes. The SPR sensorgram showed the profile of SCOOP12 peptides at gradient concentrations flowing through MIK2 ECD peptides immobilized on the chip (Fig. 4e). The analysis of the binding at equilibrium for SCOOP12 and MIK2 ECD indicated a calculated K d of 1.78 µM (Fig. 4f). A similar analysis of the SPR sensorgram using the MIK2 ECD sensor chip flowing through with SCOOP12 SS/AA peptides (Fig. 4g) indicated nearly no binding between MIK2 ECD and SCOOP12 SS/AA with a K d of 3481 µM ( Fig. 4h). Together, our data indicate that the extracellular domain of MIK2 directly binds to SCOOP12, and the conserved SxS motif of SCOOP12 is critical for its binding to MIK2. It is noted that the binding constants calculated from the in vitro binding assays are significantly different from the effective concentrations estimated from the physiological experiments. The difference might be due to conditions between in vivo and in vitro assays and the involvement of other cellular components in regulating SCOOP12-MIK2 binding in vivo 40 .
BAK1 and SERK4 are coreceptors of MIK2 in mediating SCOOP-triggered immunity. Ligand perception by LRR-RK receptors often recruits the BAK1/SERK4 family coreceptors to activate downstream signaling 8,9 . In line with a previous study 23 , SCOOP10 B -and 12-induced root growth inhibition was partially compromised in bak1-4 compared to WT (Fig. 5a,   . f SCOOP12-induced ROS production is blocked in mik2 mutants. One-week-old seedlings were treated with 100 nM SCOOP10 B or SCOOP12, and ROS production was calculated as total RLU (***P < 0.001, two-tailed Student's t-test, n = 3). g SCOOP10 B and SCOOP12-induced MAPK activation is blocked in mik2-1. Ten-day-old seedlings were treated with or without 1 µM SCOOP12. The MAPK activation was analyzed by immunoblotting (IB) with α-pERK1/2 antibodies. The protein loading is shown by CBB staining for RBC. h The mik2-1 mutant blocks SCOOP-induced root meristem distortion. Confocal images of root tips of 5-day-old seedlings with or without treatment of 1 µM peptides for 48 h followed by PI staining. Arrowheads indicate the boundary of the root meristem and the elongation zone. Scale bar, 50 µm. i SCOOP-induced defense gene expression is blocked in mik2 mutants. Ten-day-old seedlings were treated with or without 1 µM SCOOP10 B , 12, or 13 for 1 h and subjected to RT-qPCR analysis. Expression of genes was normalized to that of UBQ10 (*P < 0.05, **P < 0.01, n.s., no significant differences, two-tailed Student's t-test, n = 4). the SCOOP-mediated root growth inhibition was alleviated in the bak1-5/serk4 mutant (Fig. 5a, b). Similarly, SCOOP12induced ROS production was reduced in bak1-4 and abolished in bak1-5/serk4 (Fig. 5c). In addition, SCOOP12-or 13-induced MAPK activation was reduced in bak1-4 and substantially blocked in bak1-5/serk4 (Fig. 5d, Supplementary Fig. 6a). To examine whether the SCOOP perception induces MIK2 and BAK1 association, we performed co-immunoprecipitation (co-IP) assays in the mik2-1 mutant transformed with GFP-tagged MIK2 under the control of its native promoter (pMIK2::MIK2-GFP/mik2) with α-BAK1 antibodies. The association between MIK2 and BAK1 was barely detectable in the absence of peptide treatment but was stimulated upon the treatment of SCOOP10 B or 12 (Fig. 5e). SCOOP10 B -and 12-induced MIK2-BAK1 association was also observed in protoplasts co-expressing the hemagglutinin (HA)tagged MIK2 (MIK2-HA) and FLAG epitope-tagged BAK1 (BAK1-FLAG) ( Supplementary Fig. 6b). SERK4 was also associated with MIK2 after SCOOP peptide treatments (Fig. 5f).
To test whether the extracellular domains of MIK2 and BAK1 are sufficient to form a SCOOP12-induced complex in vitro, we performed a gel filtration assay of BAK1 ECD and MIK2 ECD purified from insect cells in the presence of SCOOP12 or SCOOP12 SS/AA peptides. The results show that MIK2 ECD and BAK1 ECD proteins co-migrated in the presence of SCOOP12 (Fig. 5g, h), indicating that SCOOP12 induces the dimerization of MIK2 ECD and BAK1 ECD . The protein complex was eluted mainly at the position corresponding to a size of a monomeric MIK2 ECD -BAK1 ECD (~120 kDa) (Fig. 5g, h), suggesting that SCOOP12 may not induce the homodimerization of the MIK2 ECD -BAK1 ECD complex. In contrast, SCOOP12 SS/AA , which cannot bind to MIK2 ECD , was unable to induce the dimerization between MIK2 ECD and BAK1 ECD (Fig. 5i, j), indicating that the SCOOP12-MIK2 binding is required for the MIK2-BAK1 interaction.  the phosphorylation of BIK1, as indicated by a protein band mobility shift in the immunoblot (Fig. 5k). SCOOP12-induced ROS burst and SCOOP10 B -induced growth inhibition were significantly compromised in the bik1/pbl1 mutant (Fig. 5c, l, m). Yet, the bik1/pbl1 mutant did not completely block SCOOP response, likely due to the functional redundancy of RLCKs 41,42 or the existence of an RLCK-independent pathway downstream of SCOOP-MIK2. Nevertheless, the MIK2-BAK1 receptor complex activation by SCOOPs triggers phosphorylation of BIK1 and PBL1 in transducing signaling to downstream events. The activation of NADPH oxidase RBOHD by BIK1 is required for PRRinduced ROS production 43,44 . Likewise, SCOOP12-induced ROS production was abolished in the rbohD mutant ( Supplementary   Fig. 6c). BAK1 and BIK1/PBL1 are also required for the MIK2dependent response triggered by a Fusarium elicitor 32 . Collectively, our results indicate that the SCOOP-MIK2-BAK1 receptorsome activates the conserved signaling pathways shared by other LRR-RK containing PRR complexes.
Microbial SCOOP-LIKE peptides trigger a MIK2-dependent immune response. A recent report showed that the mik2 mutants were insensitive to an unidentified proteinous elicitor isolated from several Fusarium spp., suggesting that MIK2 may perceive a peptide elicitor from Fusarium 32 . This prompted us to examine whether SCOOP-LIKE (SCOOPL) peptides are encoded in the . c SCOOP12-induced ROS is compromised in bak1-4, bak1-5/serk4, and bik1/pbl1. One-week-old seedlings were treated with 100 nM SCOOP12, and ROS production was calculated as total RLUs. d SCOOP12-induced MAPK activation is compromised in bak1-4 and bak1-5/serk4. The assay was done as in Fig. 3g. e SCOOPs induce MIK2-BAK1 association. Leaves of pMIK2::MIK2-GFP/mik2-1 plants were treated with or without 1 μM SCOOPs for 30 min. Total proteins were subjected for immunoprecipitation (IP) with α-GFP agarose beads, and the immunoprecipitated proteins were detected with α-BAK1 or α-GFP antibodies (top two). The input controls before IP are shown on the bottom two panels. f SCOOPs induce MIK2-SERK4 association. Protoplasts expressing MIK2-HA and SERK4-FLAG, or a control vector (Ctrl) were treated with or without 1 μM SCOOPs for 15 min. IP and IB were performed as (e). g, h SCOOP12 induces MIK2 ECD -BAK1 ECD interaction. Gel filtration chromatography analysis using MIK2 ECD and BAK1 ECD shows the elution profiles in the presence (red) or absence (black) of 0.1 mg SCOOP12 (g). SDS-PAGE and CBB staining show the eluted fractions (top panel with SCOOP12; bottom panel without SCOOP12). Stars indicate the eluted BAK1 ECD complexing with MIK2 ECD (h). i, j SCOOP12 SS/AA does not induce MIK2 ECD -BAK1 ECD interaction. Experiments were performed as in (g) and (h). k SCOOPs induce BIK1 phosphorylation. Protoplasts expressing BIK1-HA were treated with 1 μM SCOOP10 B or SCOOP12 for 15 min. Proteins were subjected to IB using α-HA antibodies (top), and CBB staining for RBC as loading controls (bottom). Phosphorylated BIK1 (pBIK1) was indicated as a band mobility shift. l, m SCOOP10 B -triggered root growth inhibition is compromised in bik1/pbl1. Images of 10-day-old seedlings with or without 1 µM SCOOP10 B . Scale bar, 4 mm (l). Quantification of root length (m). The box plots in (b), (c), and (m) show the first and third quartiles as bounds of box, split by the medians (lines), with whiskers extending 1.5-fold interquartile range beyond the box, and minima and maxima as error bar. A significant difference is shown between different letters (P < 0.05, one-way ANOVA followed by Tukey's test, n = 15, b, 5, c, and 8, m). Experiments in (a-d), (l), and (m) were repeated three times and (e-k) twice with similar results.
genomes of Fusarium spp. We blast-searched the genome of F. graminearum, a Fusarium strain with a well-annotated genome sequence 45 , using SCOOP10 B and 12 as queries. A 21-amino-acid peptide sequence in the N-terminus of an uncharacterized protein (FGSG_07177) showed a high similarity to SCOOP12 (Fig. 6a,  Supplementary Fig. 7a-c). FGSG_07177 is predicted as a transcription regulator with a GAL4-like DNA-binding domain ( Supplementary Fig. 7b). Structural modeling indicates that the SCOOPL motif is located on the surface of FGSG_07177 (Fig. 6b). Blast-searching showed that the SCOOPL sequence is highly conserved among all Fusarium species surveyed, and all searched SCOOPLs localize in the N-terminus of FGSG_07177 orthologs, which appear to be a single copy gene in different Fusarium species (Fig. 6a, Supplementary Fig. 7c). Alignment of 22 SCOOPL sequences in FGSG_07177 orthologs from different Fusarium species revealed seven groups, most of which, except for F. pseudograminearum, contain an SxS motif as Arabidopsis SCOOPs (Fig. 6a, Supplementary Fig. 7c). The second conserved serine residue of F. pseudograminearum SCOOPL (FpgSCOOPL) is substituted by a proline (Fig. 6a, Supplementary Fig. 7c). RT-qPCR analyses indicated that the FGSG_07177 homologous gene (FOXB_11846) in Fo5176, a virulent strain on Arabidopsis and belonging to F. oxysporum f. sp. conglutinans 46 , is expressed, and its expression was upregulated at 48 h upon fungal inoculation in Arabidopsis roots (Supplementary Fig. 7d).
To examine whether Fusarium SCOOPL peptides trigger immune responses as Arabidopsis SCOOPs, we synthesized five Fusarium SCOOPL peptides from F. oxysporum f. sp. FgSCOOPL peptide is highlighted in magenta, the C-terminus in yellow, and the rest in green. c Fusarium SCOOPL peptides, except FpgSCOOPL, inhibit root growth in a MIK2-dependent manner. Images of 10-day-old WT and mik2-1 seedlings with or without 1 µM Fusarium SCOOPL peptides. Scale bar, 4 mm (left). Quantification of seedling root length (right) (***P < 0.001, n.s., no significant differences, two-tailed Student's t-test, n = 9, 9,8,9,9,10,9,9,8,9,10,9 for plots from left to right). d Fusarium SCOOPLs, except FpgSCOOPL, induce ROS production. WT seedlings were treated with or without 1 µM peptides, and ROS was calculated as total RLUs. Different letters indicate a significant difference with others (P < 0.05, one-way ANOVA followed by Tukey's test, n = 4). e FocSCOOPL triggers the cytosolic Ca 2+ increase. p35S::Aequorin seedlings were treated with or without 1 µM FocSCOOPL or FpgSCOOPL. Cytosolic Ca 2+ concentration was indicated as means of RLU. The data are shown as mean ± SE (n = 9 and 6 for FocSCOOPL and FpgSCOOPL). f FocSCOOPL-induced MAPK activation is blocked in mik2-1. Ten-day-old seedlings were treated with or without 1 µM FocSCOOPL. g Fusarium SCOOPLs induce pMYB51::GUS expression in roots. Images of root tips of 1-week-old pMYB51::GUS seedlings upon treatments with or without 1 µM SCOOPLs for 3 h and subjected to GUS staining. Scale bar, 1 mm. h, i FocSCOOPL-induced ROS production is blocked in mik2 (h), and in bak1-4, bak1-5/serk4, bik1/pbl1, and rbohD mutants (i). One-week-old seedlings were treated with 1 µM FocSCOOPL, and ROS production was calculated as total RLUs. Different letters indicate a significant difference with others (P < 0.05, one-way ANOVA followed by Tukey's test, n = 6, 6, 4 for plots from left to right in (h), n = 3 in (i)). j FocSCOOPL induces MIK2-BAK1 association. The experiment was performed as in Fig. 5f. Box plots in c, d, h, and i show the first and third quartiles as bounds of box, split by the medians (lines), with whiskers extending 1.5-fold interquartile range beyond the box, and minima and maxima as error bar. Experiments in (c-j) were repeated three times with similar results.

Discussion
The Arabidopsis genome encodes more than 1000 putative secreted peptide genes, most of which have unknown functions 34 . Similarly, plants have also evolved a large number of RKs, with only a few having defined ligands and functions 3 . In this study, we report that LRR-RK MIK2 recognizes multiple plant endogenous peptides of SCOOP family members, leading to a series of PTI responses, including cytosolic Ca 2+ influx, ROS burst, MAPK activation, ethylene production, and defense-related gene expression. SCOOP12 directly binds to the extracellular LRR domain of MIK2 in vivo and in vitro, indicating that MIK2 is a bona fide receptor of SCOOPs. The SxS signature motif is essential for SCOOP functions and MIK2 binding. Perception of SCOOPs by MIK2 induces the heterodimerization of MIK2 with BAK1 and SERK4, and BAK1/SERK4 are required for SCOOPtriggered responses, indicating that BAK1/SERK4 are coreceptors of MIK2 in perceiving SCOOPs. SCOOPs have been identified as secreted peptides specific to the Brassicaceae family 23 . Interestingly, the SCOOP active motif was also found in a conserved yet uncharacterized protein ubiquitously present in fungal Fusarium spp. and bacterial Comamonadaceae. Both Fusarium and Comamonadaceae SCOOP-LIKE peptides with the SxS signature activate the MIK2-dependent immune responses. Thus, our data reveal a dual role of MIK2 in perceiving the conserved SCOOP signature motif from plants and microbes in triggering plant immunity ( Supplementary Fig. 10a). Our study is consistent with an accompanying report in which MIK2 was identified as a receptor of plant SCOOPs and mediated response to Fusarium SCOOP-like peptides 47 .
Plant plasma membrane-resident RKs perceive diverse exogenous and endogenous signals via the extracellular domain and activate intracellular responses through the cytoplasmic kinase domain [1][2][3] . The chimeric receptors with the swapped extracellular and intracellular domains between different RKs have been used to study the specificity of signal perception and signaling activation [48][49][50][51][52] . The extracellular domains of RKs determine the ligand-binding specificity 50,51,53 . The intracellular kinase domains of different RKs also trigger specific responses 48 . For example, the chimeric receptor of the EFR extracellular domain and WALL-ASSOCIATED KINASE 1 (WAK1) cytoplasmic kinase domain triggers defense responses that are activated by oligogalacturonides (OGs), the proposed ligand of WAK1 46 . We show that the chimeric receptor of RLK7 ECD -MIK2 TK or PEPR1 ECD -MIK2 TK activates some immune responses that are not observed upon RLK7 or PEPR1 activation by the corresponding ligands. More importantly, the immune responses triggered by RLK7 ECD -MIK2 TK or PEPR1 ECD -MIK2 TK largely overlapped with those activated by SCOOPs, the ligands of MIK2. For example, RLK7 activation by ligand PIP1 moderately inhibits root growth and weakly induces ROS production 19 . In contrast, the RLK7 ECD -MIK2 TK activation by PIP1 causes severe growth inhibition and robust ROS production, similar to SCOOP treatments. PIP1activated responses are BIK1-independent 19 , whereas SCOOPs induce BIK1 phosphorylation and BIK1-dependent responses. Thus, upon ligand perception by the extracellular domain, the cytosolic kinase domain of RKs activates some convergent and unique signaling events, likely through differential phosphorylation events and interaction with various partners.
Fourteen SCOOPs have been previously identified in Arabidopsis, and SCOOP12 can activate defense response and regulate root elongation 23 . Our study extends the SCOOP family to 23 members, and all tested ones with the conserved SxS motif activate plant immune response. It remains unknown why plants have evolved so many SCOOPs and the functional specificity for different SCOOPs. Notably, SCOOP genes show different expression patterns, with some highly expressed in leaves and some in roots. SCOOP12, the most studied SCOOP, is highly expressed in roots (Fig. 2b), and was indicated to regulate root development and resistance to necrogenic bacterium Erwinia amylovora and necrotrophic fungus Alternaria brassicola 23 . However, MIK2 is ubiquitously expressed in different plant tissues ( Supplementary Fig. 4e, and http://bar.utoronto.ca/efp/cgibin/efpWeb.cgi) 28 , and regulates many plant processes, including cell wall damage response, salt tolerance, and pollen tube development 24,28 . It is likely that specific SCOOPs might be recognized by MIK2 in different plant tissues to regulate distinct physiological responses. SCOOP12 triggers a robust immune response in roots, and physiological changes, such as darkened hypocotyl-root junctions and distorted meristems in roots. In addition, the mik2 mutant showed increased susceptibility to the root-invading pathogen F. oxysporum (Supplementary Fig. 8e, f) 28 . However, the enhanced susceptibility to F. oxysporum has not been observed in the scoop12 mutant, likely due to the functional redundancy of multiple SCOOPs in Arabidopsis. Alternatively, it could be attributed to SCOOPL genes in F. oxysporum.
SCOOPL sequences are also present in a wide range of fungal Fusarium spp. and bacterial Comamonadaceae. In Fusarium spp., the SCOOPL motif we identified is located in the N-terminus of a highly conserved GAL4 DNA-binding domain-containing protein with uncharacterized functions. It is possible that SCOOPL motifs are also present in some other proteins in Fusarium spp 47 . Some of these SCOOPLs are active in triggering MIK2-and BAK1/SERK4-dependent immune responses (Fig. 6c, h, i). Knock-out of the SCOOPL in Fo5176 enhanced the fungal pathogenicity in Arabidopsis (Supplementary Fig. 8c, d). Therefore, Fusarium SCOOPLs may be recognized as MAMPs by plants. It will also be interesting to determine whether Fusarium SCOOPL is the proteinous elicitor isolated from several Fusarium spp., which triggers a MIK2-dependent immunity 32 . It remains unknown whether this protein is secreted and processed during Fusarium infection and how MIK2 perceives it. Notably, elf18, an 18-amino acid synthetic peptide, is derived from the conserved bacterial translation elongation factor EF-Tu, which can be detected on the extracellular surface and secretome of bacteria [54][55][56] . Some SCOOPL peptides from Fusarium spp. and Comamonadaceae have amino acid variations in the conserved SxS motif and cannot activate the immune response. This is consistent with the observation that the SxS motif is required for plant SCOOP peptide functions and binding to MIK2. The variations may be due to the pathogens' evolutionary pressure to escape from hosts' perception of immune elicitation.
It has been reported that some pathogens and nematodes deploy mimics of plant endogenous peptides, such as CLA-VATA3/ESR (CLE), PLANT PEPTIDE CONTAINING SUL-FATED TYROSINE (PSY), and RAPID ALKALINIZATION FACTOR (RALF), to promote the pathogenicity by hijacking plant peptide-receptor signaling [57][58][59][60] . For example, RALF secreted from F. oxysporum induces plant receptor FERONIAdependent extracellular alkalization to favor fungal multiplication 57 . In contrast, similar to plant SCOOPs, microbial SCOOPLs activate the MIK2-dependent immune responses, suggesting that SCOOPLs act as MAMPs rather than virulence factors. Compared to the wide distribution of SCOOPLs in fungal Fusarium spp. and bacterial Comamonadaceae, plant SCOOPs are only present in Brassicaceae plants ( Supplementary  Fig. 4d) 23,33 . This indicates that plant SCOOPs may have evolved later than microbial SCOOPLs. In addition, gene duplications of SCOOPs are common in Brassicaceae species 23 , an indicative for highly evolved genes. For example, twelve SCOOPs exist as tandem repeats on Arabidopsis chromosome 5 23 . Thus, plant SCOOPs may have convergently evolved to mimic microbial SCOOPLs and amplify SCOOPL-triggered immunity. Phylogenetic analysis indicates that peptide motifs of Arabidopsis SCOOPs, Fusarium and Comamonadaceae SCOOPLs might have evolved independently (Supplementary Fig. 10b). Moreover, plant SCOOPs are derived from small peptide precursor proteins, whereas Fusarium and Comamonadaceae SCOOPLs reside in proteins belonging to distinct families. The divergence of protein families harboring SCOOP/SCOOPL suggests that SCOOPs and SCOOPLs might have evolved convergently but unlikely by horizontal gene transfers 61 .
Plasmid construction and generation of transgenic plants. The BAK1 (AT4G33430), SERK4 (AT2G13790), BIK1 (AT2G39660), and MIK2 (AT4G08850) coding sequences (CDSs) were amplified by PCR from Col-0 cDNA using the primers containing BamHI at the 5′ end and StuI at the 3′ end. The PCR products were digested by BamHI and StuI, followed by ligation into the pHBT-FLAG or pHBT-HA vector to generate pHBT-35S::BAK1-FLAG, pHBT-35S::SERK4-FLAG, pHBT-35S::BIK1-HA, pHBT-35S::MIK2-HA, and pHBT-35S::MIK2-GFP constructs 24,41,62 . MIK2 ECD (residues 1-707) and BAK1 ECD (residues 1-220) were amplified by PCR from Col-0 cDNA and cloned into pFastBac-1 vector with a C-terminal 6× HIS tag 25,62 . The fusion protein vectors carrying MIK2 ECD or BAK1 ECD were used for insect cell expression. To obtain the binary vector pCAMBIA1300-35S::RLK7, a 2907-bp RLK7 CDS was PCR-amplified from WT cDNA using gene-specific primers with KpnI and SalI at the 5′ and 3′ ends, respectively, followed by KpnI and SalI digestion and ligation into the pCAM-BIA1300 vector. To generate the binary vector pCAMBIA1300 carrying the RLK7 ECD -MIK2 TK chimeric receptor gene, an 1824-bp fragment encoding RLK7 ECD was PCR-amplified from WT cDNA using a gene-specific forward primer with KpnI at the 5′ end and a gene-specific reverse primer with 8-bp overlapping sequence from the 5′ end of MIK2 TK , and a 1011-bp fragment encoding MIK2 TK using a gene-specific forward primer with 8-bp overlapping sequence from the 3′ end of RLK7 ECD and a gene-specific reverse primer with SalI at the 5′ end. RLK7 ECD was fused with MIK2 TK and inserted into the pUC19 vector using infusion recombinant enzymes (Clontech). After digestion with KpnI and SalI, the RLK7 ECD -MIK2 TK chimeric receptor gene was inserted into a pCAMBIA1300 vector with the CaMV 35 S promoter and the HA tag at the 3′ end to generate pCAMBIA1300-p35S::RLK7 ECD -MIK2 TK . A similar strategy was used to generate pCAMBIA1300-35S::PEPR1 ECD -MIK2 TK . A 2307-bp fragment encoding PEPR1 ECD was amplified for fusion with MIK2 TK . To obtain pHBT-35S::MIK2-HA constructs, the MIK2 CDS was PCR-amplified from WT cDNA using gene-specific primers with BamHI and StuI at 5′ and 3′ ends, respectively, followed by BamHI and StuI digestion and ligation into the pHBT vector with HA sequence at the 3′ end. To generate the binary vector pCAMBIA1300-pMIK2::MIK2-GFP, a 2000-bp promoter sequence upstream of the start codon of MIK2 was PCR-amplified and subcloned into pHBT-35S::MIK2-GFP between XhoI and BamHI sites. The pMIK2::MIK2-GFP-NOS fragment was further PCR amplified and inserted into pCAMBIA1300 between EcoRI and SalI using in-fusion recombinant enzymes. To generate the binary vector pCAMBIA-pSCOOP12::SCOOP12-GFP, a 2481-bp fragment containing the SCOOP12 promoter and coding sequence was PCR-amplified using Col-0 genomic DNA as a template. The fragment was further inserted into NcoI-digested pCAMBIA1300-GFP using in-fusion recombinant enzymes. The primers used for cloning and sequencing were described in Supplementary Data 3, and the Sanger-sequencing verified all insertions in different vectors.
The binary vectors were introduced into Agrobacterium tumefaciens strain GV3101 for the floral dipping method-based Arabidopsis transformation. Transgenic plants were selected with 20 µg/mL hygromycin B. Multiple transgenic lines in the T 1 generation were analyzed by immunoblotting for protein expression. Two lines with a 3:1 segregation ratio for hygromycin resistance in the T 2 generation were selected to obtain homozygous seeds.
To generate a vector for recombinant protein expression of SCOOP12 in Escherichia coli, a 147-bp fragment encoding SCOOP12 lacking the N-terminal signal peptide was PCR-amplified from pCAMBIA-pSCOOP12::SCOOP12-GFP using gene-specific primers with BamHI and HindIII at 5′ and 3′ ends, respectively, followed by BamHI and HindIII digestion and ligation into pET28a-Avi-HIS 6 -SUMO 64 to create pET28a-Avi-HIS 6 -SUMO-SCOOP12.
Peptide synthesis. PIP1, Pep1, flg22, and all non-labeled SCOOPs and SCOOP-LIKE peptides were synthesized at ChinaPeptides (Shanghai, China). TAMRAlabeled SCOOP12 peptides were synthesized at Biomatik (Delaware, USA). The sequences of synthesized peptides were listed in Supplementary Data 4.
Root growth assay. Cold stratified seeds were surface-sterilized with 70% (v/v) ethanol for 5 min and were sown on ½MS plates with or without peptides at indicated concentrations. Ten-day-old seedlings grown on plates vertically in a growth chamber were photographed, and the root lengths of seedlings were measured using Image J (http://rsb.info.nih.gov/ij/).
RNA sequencing (RNA-Seq). Ten-day-old WT Col-0, mik2-1, and RLK ECD -MIK TK (L16) seedlings grown on ½MS plates were incubated in 1 mL of liquid ½MS medium overnight. Seedlings were then treated with or without 1 μM PIP1 or SCOOP12 for 1 or 6 h and harvested for RNA isolation. Total RNAs (5 μg) from two biological replicates were pooled for cDNA library construction. cDNA library preparation and sequencing were carried out on an Illumina HiSeq 4000 platform with 150-nucleotide pair-end reads in LC-BIO (Hangzhou, China). Total reads were mapped to the Arabidopsis genome (TAIR10; www.arabidopsis.org) with Hisat, and the read counts for every gene were generated with StringTie. The expression levels of genes were calculated by quantifying the cDNA fragments per kilobase of transcript per million fragments mapped (FPKM). Differentially expressed genes (DEGs) between different treatments were defined by fold change of read counts at 1 and 6 h compared to that at 0 h (differential expression analysis using edgeR, fold change ≥ 2 with p-value ≤ 0.01). DEGs were listed in Supplementary Data 2.
RNA isolation and reverse transcription-quantitative polymerase chain reactions (RT-qPCR) analysis. For detection of gene expression in Arabidopsis, the total RNA was extracted from 10-day-old seedlings grown on ½MS plates, or from rosette leaves or inflorescences of 4-or 7-week-old soil-grown plants using TRIzol reagent (Invitrogen). For detecting gene expression in F. oxysporum during infections, 3-week-old Arabidopsis seedlings were gently taken out of soil, and roots were inoculated with 1 × 10 7 /mL F. oxysporum spores after washing with sterile water to clear off soil chunks. After incubating on sterile wet papers for 0-48 h, F. oxysporum-infected Arabidopsis roots were collected for total RNA extraction using TRIzol reagent (Invitrogen). One microgram of total RNA was reverse-transcribed to synthesize the first-strand cDNA with M-MuLV Reverse Transcriptases (Thermo Fisher Scientific) and oligo(dT) primers following by RNase-free DNase I (Thermo Fisher Scientific) treatment. RT-qPCR analyses were performed on a QuantStudio™ 3 Real-Time PCR Detection System (Thermo Fisher Scientific) using Faster Universal SYBR ® Green Master (Roche) and gene-specific primers. The expression of each gene in Arabidopsis and F. oxysporum was normalized to the expression of Arabidopsis UBQ10 and F. oxysporum GAPDH, respectively. The primers used for RT-qPCR were listed in Supplementary Data 3.
ROS assay. ROS burst was determined by a luminol-based assay. Five one-weekold seedlings grown on ½MS plates were incubated in 200 μL ddH 2 O overnight in a 1.5 mL centrifuge tube. Then, ddH 2 O was replaced by 200 µL of reaction solution containing 50 µM of luminol, and 10 µg/mL of horseradish peroxidase (Sigma-Aldrich) supplemented with or without 100 nM or 1 μM peptide. Luminescence was measured immediately after adding the solution with a luminometer (Glomax 20/20n, Promega) with a one-second interval for 15 min. The total values of ROS production were indicated as means of the relative light units (RLUs).
Measurement of cytosolic Ca 2+ concentration. One-week-old seedlings expressing p35S::Aequorin grown vertically on ½MS plates were put into a 1. Measurement of ethylene production. Twelve leaves of 4-week-old plants grown on soil were excised into leaf disks of 0.25 cm 2 , followed by overnight incubation in a 25 mL glass vial with 2 mL of ddH 2 O for recovery. Then, ddH 2 O was replaced by 1 mL of peptides at 1 µM, and the vials were capped immediately with a rubber stopper and incubated at 23°C with gentle agitation for 4 h. One mL of the vial headspace was injected into a TRACE TM 1310 Gas Chromatograph (Thermo Fisher Scientific) with FID supported by Chromeleon 7 for quantitation.
Microscopy assays. For the observation of propidium iodide (PI) stained roots, 5day-old seedlings grown on ½MS plates were mounted in ddH 2 O containing 10 μM PI for 20 min before imaged with the Leica SP8 confocal microscope. Images were captured at 543 nm laser excitation and 578-700 nm emission. For the observation of TAMRA-SCOOP12 labeled roots and protoplasts, 5-day-old seedlings grown on ½MS plates or protoplasts isolated from leaves of 4-week-old soilgrown plants were treated with 100 nM TAMRA-SCOOP12 with or without 1 μM SCOOP12 or flg22 for 5 min in liquid ½MS or W5 solution (154 mM NaCl, 125 mM CaCl 2 , 5 mM KCl, 2 mM MES, pH 5.7), followed by washing with ½MS or W5 solution for three times before imaged with the Leica SP8 confocal microscope. Images were captured at 552 nm laser excitation and 570-620 nm emission. To observe SCOOP12-GFP in Arabidopsis pSCOOP12::SCOOP12-GFP/scoop12 transgenic plants, detached leaves were imaged under the Leica SP8 confocal microscope directly or 3 min after 5% sodium chloride (NaCl) treatment. Images were captured at 472 nm laser excitation and 493-547 nm emission. For all the observations, the pinhole was set at one Airy unit, and the imaging processing was carried out by using the Leica Application Suite X (LAS X) software.
In vitro protein cleavage assay. Recombinant HIS-SCOOP12 proteins were expressed in E. coli BL21 strain with an isopropyl β-D-1-thiogalactopyranoside (IPTG)-inducible system and purified with Ni-NTA agarose (QIAGEN, #166048536) in buffer containing 50 mM Tris-HCl pH 7.5 and 300 mM NaCl. For in vitro protein cleavage assay, 5 μg of purified HIS-SCOOP12 proteins were incubated with 10 μL protein extracts isolated from 10-day-old Arabidopsis WT seedlings or protein extraction buffer (50 mM HEPES, pH 7.4, 10 mM EDTA, and 1× protease inhibitor cocktail, Sigma) following a published protocol 36 for 5 h at room temperature with rotation. The reaction was quenched by adding SDS loading buffer followed by separation on 15% (v/v) SDS-PAGE and subjected to immunoblotting (IB) with α-HIS antibodies (1: 2000; Roche, Cat #11965085001) or Coomassie Brilliant Blue (CBB) staining. The processed SCOOP12 bands were cut from SDS-PAGE for LC-MS/MS analysis.
LC-MS/MS analysis. The processed SCOOP12 bands were in-gel digested following a published protocol 66 , except that digestion was conducted with Lys-C (Promega, Madison, WI, USA). The digested peptides were desalted using micro ZipTip mini-reverse phase (Millipore). In brief, after wetting (with 50% acetonitrile) and equilibrating a ZipTip (with 0.1% formic acid), the peptides were bound to C-18 material. A subsequent washing step with 0.1% formic acid was followed by a final elution with 60% acetonitrile and 0.1% formic acid. The samples were lyophilized to dryness using a labconco speedvac (Centrivap, Labconco Inc., USA). The resulting peptides were resuspended in 0.1% formic acid for mass spectrometric analysis. The mass spectrometry data acquisition was performed on an EASY-nLC 1200 ultraperformance liquid chromatography system (Thermo Scientific Inc., San Jose, CA) connected to an Orbitrap Q-Exactive HF instrument equipped with a nano-electrospray source (Thermo Scientific Inc., San Jose, CA) 67 . The peptide samples were loaded to a C18 trapping column (75 μm i.d. × 2 cm, Acclaim PepMap® 100 particles with 3 μm size and 100 Å pores) and then eluted using a C18 analytical column (75 μm i.d. × 25 cm, 2 μm particles with 100 Å pore size). The flow rate was set at 250 nL/min with solvent A (0.1% formic acid in water) and solvent B (0.1% formic acid and 99.9% acetonitrile) as the mobile phases. MS/MS analysis was carried out in positive mode applying data-dependent acquisition. Survey full-scan MS spectra scan range was 375-1575 m/z, with a resolution of 60,000 at 200 m/z. The precursor isolation window was set at 4 m/z, the automatic gain control (AGC) was 3e6, and the maximum injection time (IT) was 80 ms. The MS/MS resolution was 15,000, and the mass range was 200-2000 m/z. The minimum AGC target was 8.0e3, and the IT was 55 ms. The normalized collision energy was set to be 28 with apex trigger of top 7 precursor ions. The tandem mass spectra were extracted from the Xcalibur .raw files and converted into mgf files using Proteome Discoverer 2.4 (Thermo Fisher Scientific Inc., San Jose, CA). All MS/MS samples were analyzed using Mascot Daemon (Matrix Science, London, UK; version 2.4.0). Mascot was set up to search the Arabidopsis TAIR10 database assuming the digestion enzyme semi-Lys C. Decoy database searching was enabled. Carbamidomethyl of cysteine was specified in Mascot as a fixed modification. The raw MS/MS spectra were extracted and exported from the raw data file using Xcalibur Qualbrower and annotated manually with the assistance of Mascot and MS Product from ProteinProspector.
Protein expression and purification. The ECD domains of MIK2 (MIK2 ECD ) and BAK1 (BAK1 ECD ) fused with a 6 × HIS tag at the C-terminus were expressed using the Bac-to-Bac baculovirus expression system (Invitrogen) in High Five cells at 22°C 25 . One liter of cells (1.8 × 10 6 cells per mL cultured in the medium from Expression Systems) was infected with 20 mL baculovirus and the media were harvested after 48 h of infection. The secreted MIK2 ECD -6 × HIS and BAK1 ECD -6 × HIS proteins were purified using Ni-NTA (Novagen) and size-exclusion chromatography (Hiload 200, GE Healthcare) in buffer containing 10 mM Bis-Tris pH 6.0 and 100 mM NaCl.
Isothermal titration calorimetry (ITC) assays. The binding affinities of MIK2 ECD with SCOOP12 or SCOOP12 SS/AA peptides were measured on a MicroCalorimeter ITC200 (Microcal LLC) at 25°C 25 . All protein samples used in the ITC assay were dialyzed in the buffer containing 200 mM HEPES, pH 7.0, and 100 mM NaCl. Protein concentration was determined by absorbance spectroscopy at 280 nm. SCOOP12 or SCOOP12 SS/AA peptides (600 μM) were injected (1.5 μL per injection) into the stirred calorimeter cells containing 30 μM MIK2 ECD with 20 × 2 μL at 2.5-min intervals. ITC data were analyzed using MicroCal Origin 7.0.
Surface plasmon resonance (SPR) assays. The binding kinetics and affinities of MIK2 ECD with SCOOP12 or SCOOP12 SS/AA peptides were assessed on a Biacore T200 instrument (GE Healthcare, Pittsburgh, PA) with CM5 sensor chips. The MIK2 ECD proteins were exchanged to 10 mM sodium acetate (pH 5.5), and the peptides used for SPR were dissolved in HBS-EP+ (10 mM HEPES, 150 mM NaCl, 3 mM EDTA, and 0.05% [v/v] surfactant P20 (GE Healthcare)). About 4700 response units of MIK2 ECD proteins were immobilized on a CM5 chip, and a blank channel was used as a negative control. SCOOP12 or SCOOP12 SS/AA peptides were diluted into indicated concentrations and injected at a flow rate of 30 μL min −1 for 2 min, followed by dissociation for 5 min. After dissociation, 5 mM NaOH was injected for 30 s to remove any non-covalently bound proteins from the chip surface. The binding kinetics was analyzed with the software Biaevaluation ® Version 4.1 using the 1:1 Langmuir binding model.
Gel filtration assays. The purified MIK2 ECD proteins (0.5 mg for each) and BAK1 ECD proteins (0.2 mg for each) were incubated together with SCOOP12 or SCOOP12 SS/AA (0.1 mg for each) in a buffer containing 10 mM Bis-Tris, pH 6.0, 100 mM NaCl on ice for 1 h. Then, each of the mixtures was separated by gel filtration (Hiload 200, GE Healthcare) with a flow rate of 0.8 mL/min and an injection volume of 40 mL at 4°C, and the peak fractions were separated on SDS-PAGE followed by Coomassie blue (CBB) staining.
Fusarium oxysporum disease assay. F. oxysporum strain Fo5176 was grown on potato dextrose agar plates (Difco) for 4 d at 28°C. The hyphae of Fo5176 were inoculated in potato dextrose broth (Difco) and incubated in a shaker (120 rpm) at 28°C for 5 d. The spores were collected by filtering through Miracloth (Millipore) followed by centrifugation at 3000 × g for 15 min, resuspended with sterile water, and adjusted to a final concentration of 10 7 spores/mL. Three-week-old Arabidopsis seedlings were gently taken out of the soil, and roots were washed with sterile water to clear off soil chunks, blotted using paper towels, immersed in Fo5176 spore suspension or water (mock) for 5 min, and then planted back in pots with pre-wet fresh soil. The inoculated seedlings were covered with a dome for two days and incubated at 28°C, 65% humidity, and 100 μE m −2 s −1 light with a 12-h light/12-h dark photoperiod. The disease severity was evaluated using the percentage of chlorotic leaves in each plant 68 .
Phylogenetic analysis and graphical illustration. Protein sequences were retrieved from the NCBI database. Multiple sequence alignments were generated with ClustalW 69 and displayed in ESPript3 (espript.ibcp.fr). The consensus of sequences was analyzed by WebLogo (http://weblogo.berkeley.edu/logo.cgi). Phylogenetic analysis was carried out using MEGAX with the neighbor-joining method with 1000 bootstrap replicates 70 . The phylogenetic tree was visualized in an interactive tree of life (iTOL, https://itol.embl.de/). Homology modeling and visualization of F. graminearum protein FGSG_07177 were performed with Pymol (The PyMOL Molecular Graphics System, Version 1.2r3pre, Schrödinger, LLC.) with the centromere DNA-binding protein complex CBF3 subunit B (PDB-ID, c6f07B) as the template. Three-dimensional structure of SCOOP12 was predicted by I-TASSER (https://zhanglab.ccmb.med.umich.edu/I-TASSER/).

Statistical analysis.
Statistical analyses were all done in Excel with built-in formulas. The p values were calculated with two-tailed unpaired Student's t-test analysis for binary comparison or one-way ANOVA and Tukey's post hoc honest significance test to compare more than two genotypes or treatments. The measurements shown in box plots display the first and third quartiles and split by medians (center lines), with whiskers extending to 1.5-fold the interquartile range from the 25th and 75th percentiles. Dots represent outliers. Asterisks illustrate the p values: ***P < 0.001, **P < 0.01, and *P < 0.05.