Two distinct modes of action of molecular glues in the plant hormone co-receptor COI1-JAZ system

Summary The plant hormone (3R, 7S)-jasmonoyl-L-isoleucine ((3R, 7S)-JA-Ile) is perceived by the COI1-JAZ co-receptor in Arabidopsis thaliana, leading to the activation of gene expression for plant defense responses, growth, development, and other processes. Therefore, understanding the interaction between the COI1-JAZ co-receptor and its ligands is essential for the development of COI1-JAZ agonists and antagonists as potent chemical tools for regulating (3R, 7S)-JA-Ile signaling. This study demonstrated that COI1-JAZ has two independent modes of ligand perception using a differential scanning fluorimetry (DSF) assay. (3R, 7S)-JA-Ile is perceived through a one-step model in which (3R, 7S)-JA-Ile causes protein–protein interaction between COI1 and JAZ. In contrast, coronatine (COR), a mimic of (3R, 7S)-JA-Ile, is perceived through a two-step model in which COR is first perceived by COI1 and then recruits JAZ to form the COI1-COR-JAZ complex. Our results demonstrate two distinct modes of action of molecular glues causing protein–protein interactions.


ll OPEN ACCESS
7S)-JA-Ile-JAZ complex formation and may not depend on the two-step mode.Thus, we conclude that the Arabidopsis COI1-JAZ co-receptor system has two independent modes of ligand perception.

DSF assay can demonstrate the two-step ligand perception mode of COI1-JAZ co-receptor
For the DSF assay, we prepared tag-free Arabidopsis COI1 protein from glutathione-S-transferase (GST)-fused COI1 co-expressed with Arabidopsis SKP1-like (ASK1) which improved the expression yield in Sf9 cultured insect cells. 25The GST tag was removed using the Tobacco Etch Virus (TEV) protease (Figure S1). 19First, we used COR as a positive control in the DSF assay because the direct interaction of COR with COI1 has been previously confirmed using the iTC assay. 18We examined the varying concentration of COI1 (0.1, 0.5, and 1.0 mM) and 1.0 mM condition showed a clear melting curve and was employed in the following assay (Figure S2A).The melting point of COI1 shifted to a higher temperature in the presence of COR (0.3-30 mM; Figures 2A and S3A), indicating the successful observation of COI1-COR interaction using DSF.The K d value (Figure 2B; Table 1, K d = 4.08 mM) observed through the DSF assay was four times larger than that previously reported using the iTC assay (K d = 0.96 mM). 18As shown in Figures S2B and S2C, ASK1 did not affect the result of DSF assay because the background DSF curves of ASK1 were negligible compared to those of the COI1-ASK1 complex.The addition of JAZ1/9 degron peptide (JAZP1/9, Figures 2C and S4), which contains the short degron motif necessary for the interaction with the ligand, improved the K d values (Figures 2B, 2D, S3B, and S3C) (Table 1, K d = 0.60 mM for COI1-JAZ1 and K d = 0.29 mM for COI1-JAZ9).On the other hand, no shift of melting point was observed by the addition of JAZP1/9 to COI1 (B) (3R, 7S)-JA-Ile acts as a molecular glue to induce protein-protein interaction between COI1 and JAZ.(C) COR is first perceived by COI1 and subsequently recruits JAZ to form a COI1-COR-JAZ ternary complex.
(Figure 2E).This result suggests that the conformation of COI1, which is fixed by binding to the ligand, can be further tightened by adding JAZP1/9.The observed K d values were consistent with the previously obtained K d value for COI1-JAZP1 (ca.0.11 mM). 19In contrast, no shift in the melting point of COI1 was observed upon addition of racemic jasmonic acid (Figure 1A), which was used as a negative control (Figures 2F and S5A). 10,18he addition of JAZP1/9 did not affect the shift in melting point (Figures S5B and S5C).We also examined (+)-cis-12-oxo-phytodienoic acid (cis-OPDA, Figure 1A) as a negative control, 10 and no shift was observed in the melting point of COI1 upon titration (Figure S6).These results confirmed that the DSF assay provides reliable results on the interaction of COR and COI1/COI1-JAZs and demonstrated that the two-step perception mode can be proven by methods other than the iTC.
COI1 perceives the COR-derived antagonist COR-MO via a two-step perception mode COR-MO (Figure 1A), an O-methyl oxime derivative of COR, is a potent antagonist of the COI1-JAZ co-receptor. 15The putative mode of action of COR-MO is explained by a two-step perception mode: COR-MO potentially first binds to COI1 and then interferes with the recruitment of JAZ through steric hindrance caused by the O-methyl oxime group. 18This mode of action is strongly supported by molecular dynamics calculations and subsequent docking simulations. 26However, the proposed mode of action of COR-MO has not yet been confirmed experimentally.We examined the interaction between COR-MO and COI1/COI1-JAZ using the DSF assay.The melting point of COI1 shifted to a higher temperature in the presence of COR-MO (0.3-30 mM) (Figures 3A and S7A), and the K d value for the COI1-COR-MO interaction was obtained (Figure 3B, K d = 2.65 mM).K d (COI1-COR-MO) was comparable to K d (COI1-COR) (Table 1).In contrast to COR (Figure 2B), the addition of JAZP1/9 did not affect the K d value (Figures 3B, 3C, S7B, and S7C), suggesting the inhibition of JAZ recruitment to the COI1-COR-MO complex.These results are consistent with the previously proposed two-step mode of ligand perception and demonstrate that COR-MO first binds to COI1 and inhibits JAZ access to the COI1-COR-MO complex.
Meanwhile, the addition of JAZP1/9 significantly improved the K d values (Figures 4C and 4 days, S8B and S8C, K d = 17.7 mM for COI1-(3R, 7S)-JA-Ile-JAZP1, and K d = 7.39 mM for COI1-(3R, 7S)-JA-Ile-JAZP9).The K d (COI1-(3R, 7S)-JA-Ile-JAZP1/9) values were approximately 25-30 times higher than the K d (COI1-COR-JAZP1/9) values (Figures 2B and 4D).This result is consistent with the previous results of the pull-down assay, in which COR had approximately 100 times stronger affinity for COI1-JAZ9 than (3R, 7S)-JA-Ile. 12e also performed the iTC assay to validate the result of DSF assay.The iTC assay using GST-COI1, GST-COI1/JAZP1, JAZP1, and JA-Ile afforded distinct colorimetric titration curves, respectively (Figure 5).A complex of GST-COI1, JAZP1, and JA-Ile showed the clear sigmoidal curve to afford a strong K d value of 0.15 mM (Figures 5A and 5D).JAZP1 and JA-Ile only showed the heat of dilution, suggesting no interaction between JAZP1-ligand (Figures 5B and 5D).GST-COI1 and JA-Ile showed a relatively gradual titration curve and required 6 eq. of JA-Ile for saturation to afford 11.0 mM of K d value (Figures 5C and 5D), suggesting the existence of weak non-binding-pocket interaction and the interaction is negligible in the physiological conditions. 18,27e structural factor determining the mode of ligand perception by COI1-JAZ To understand the structural factors contributing to the difference between the affinity of Arabidopsis COI1 to COR and (3R, 7S)-JA-Ile, the COI1 affinities of COR and (3R, 7S)-JA-Ile hybrid molecules were examined.COR is composed of (3R, 7S)-coronafacic acid (CFA) and (2S, 3S)coronamic acid (CMA), with CFA structurally mimicking (3R, 7S)-jasmonic acid (JA) and CMA structurally mimicking L-Ile.Therefore, (3R, 7S)coronafacyl L-Ile (CFA-Ile, Figure 6A) 28,29 and (3R, 7S)/(3R, 7R)-jasmonoyl CMA (JA-CMA, Figure 6A) were synthesized as hybrid molecules (Figure S10).Their COI1 affinities were evaluated using DSF.We could not separate (3R, 7S)-and (3R, 7R)-JA-CMA by HPLC; thus, we used a 5/95 mixture of (3R, 7S)/(3R, 7R)-JA-CMA in the DSF assay (Figure S11).As expected, CFA-Ile caused a significant shift in the melting point in the DSF assay, and the K d was calculated to be 14.1 mM (Figures 6B, 6C, and S12A; Table 1).Similar to the case of COR, the addition of JAZP1/9 improved the K d values (Figures 6C, S12B and S12C) (Table 1, K d = 2.66 mM for COI1-JAZP1 and K d = 2.95 mM for COI1-JAZP9).In contrast, there was little or no shift in the melting point upon the addition of JA-CMA (Figures 6D and S13A; Table 1).Furthermore, adding JAZP1/9 improved the K d values (Figures 6E, S13B, and S13C) (Table 1, K d = 33.5 mM for COI1-JAZP1 and K d = 24.2mM for COI1-JAZP9).In addition, (3R, 7S)-coronafacyl L-Val (CFA-Val), which is known to bind with COI1-JAZ1 co-receptor, 28,29 caused a significant shift in the melting point of COI1 and the addition of JAZP1/9 improved the K d values as anticipated (28.1 mM for COI1, 6.63 mM for COI1-JAZP1 and K d = 9.77 mM for COI1-JAZP9) (Table 1; Figure S14).These results demonstrated that the CFA moiety determines the ligand-binding mode, either one-step or two-step.To examine the role of the CFA moiety in binding, we performed docking simulations of CFA-Ile and JA-CMA with COI1.Sandwich-type interactions of the ligands with the two aromatic side chains of COI1, Phe89 and Tyr444, were observed (Figures 6F and 6G).There were no differences between the hydrogen-bonding networks of COI1-JAZ1, CFA-Ile, and JA-CMA (Figures 6F and 6H).It was estimated that the strength of the hydrophobic interactions would be much greater in CFA-Ile than in JA-CMA, considering that the cyclohexene ring of CFA was oriented face-to-face to the phenyl group of Phe89 (Figure 6G), which would contribute to the observed stronger affinity of CFA-Ile for COI1.

DISCUSSION
Various in vitro biochemical assays have been used to study the interactions between COI1-JAZ and its ligands. 20,30Most of these in vitro assays depend on the ligand-mediated formation of the COI1-ligand-JAZ ternary complex.These assays include pull-down, 10,11 yeasttwo-hybrid (Y2H), 10 surface plasmon resonance (SPR), 18,31 AlphaScreen, 32 and fluorescence anisotropy. 33In contrast, the iTC assay, which measures the heat exchange during ligand-receptor complexation, does not depend on ternary complex formation and can monitor the direct interaction between the ligand and COI1 or JAZ.However, the iTC assay requires a large amount of COI1 protein; therefore, an assay method for COI1-ligand interactions that requires a low amount of COI1 protein is key.
In this study, we used the DSF assay, which requires a smaller amount of COI1 protein than iTC, to study the direct affinity of ligands for the Arabidopsis COI1 protein.DSF assays have been successfully used to analyze the receptor interactions of other plant hormones, such as abscisic acid and strigolactones. 22,24,34Using the DSF assay, we confirmed the two-step perception of COR by the COI1-JAZ coreceptor (Figures 2A, 2B, and 2D).To the best of our knowledge, this is the first successful application of DSF in the study of COI1ligand interactions.Compared to previous results from iTC and FA assays, 18,33 the DSF assay showed a larger K d value for the COI1-JAZ co-receptor-ligand interaction, indicating a weaker evaluation of binding.Using the DSF assay, we also demonstrated the direct interaction of COR-MO, a potent antagonist of the COI1-JAZ co-receptor, 15 and COI1, confirming the previously proposed but unconfirmed mode of action of this antagonist: COR-MO first binds COI1 and then perturbs the recruitment of JAZ, possibly because of steric hindrance due to the O-methyl oxime group (Figures 3A and 3B). 26In addition, the melting point of the COI1-COR-MO complex was not affected by the addition of JAZP (Figure 3C), indicating that the DSF assay can be used to efficiently screen for COI1-JAZ antagonists by comparing the melting points of COI1-ligand complexes with and without JAZP.The DSF assay is a powerful method for examining the direct affinity of a ligand for COI1 and provides valuable information for the development of a synthetic agonist/antagonist of the COI1-JAZ co-receptor system.
Our results suggest little or no affinity of (3R, 7S)-JA-Ile for COI1 (Figure 4A).Thus, unlike COI1-COR, COI1 alone may not function as a receptor for (3R, 7S)-JA-Ile in plant cells.The wound-induced accumulation of (3R, 7S)-JA-Ile in Arabidopsis thaliana is estimated to reach approximately 0.9-1.6 mM, 18,27 and the local concentration of (3R, 7S)-JA-Ile in the nucleus, where COI1 is localized, 35 is likely to be higher.However, these concentrations may not be sufficient to facilitate the interactions between COI1 and (3R, 7S)-JA-Ile.In addition, a previous study has reported that JA-Ile-MO (Figure 1A), the JA-Ile counterpart of COR-MO, does not function as a COI1-JAZ antagonist because of its extremely low inhibitory activity. 15These previously reported results are consistent with our conclusion that COI1 alone cannot perceive (3R, 7S)-JA-Ile to form a COI1-(3R, 7S)-JA-Ile complex for the subsequent recruitment of JAZ proteins.However, our current results also raise the possibility that the little or no affinity of (3R, 7S)-JA-Ile for COI1 could be due to the complete isomerization of the active cis-isomer (3R, 7S)-JA-Ile to the inactive trans-isomer (3R, 7R)-JA-Ile (Figure 1A), and we confirmed that (3R, 7R)-JA-Ile has little affinity for COI1 (Figure S9).UPLC-MS/MS analyses revealed that 36.1% of cis-isomer (3R, 7S)-JA-Ile remained stable under DSF conditions (Figure 4B).In a previous study, 65% of (3R, 7S)-JA-Ile was isomerized to (3R, 7R)-JA-Ile after 1 h of heating at 95 C 12 ; this result is consistent with the results of the current study.However, the significant increase in K d values when JAZP1/9 was added to the COI1-(3R, 7S)-JA-Ile complex suggests that complete isomerization and deactivation did not occur during the DSF assay.Thus, we evaluated the affinity of (3R, 7S)-JA-Ile for COI1 using the DSF assay.It was concluded that (3R, 7S)-JA-Ile is perceived by the COI1-JAZ co-receptor through a one-step formation of the ternary complex COI1-(3R, 7S)-JA-Ile-JAZ.
In conclusion, our results showed the Arabidopsis COI1-JAZ co-receptor system can perceive a ligand through a one-or two-step perception mechanism.Our results suggest that the two-step perception mode for interaction with COI1-JAZ and its ligands applies to COR and COR-based ligands.In contrast, the plant hormone (3R, 7S)-JA-Ile may be perceived by the COI1-JAZ co-receptor in a one-step mode.Thus, we demonstrate two distinct modes of action of molecular glue in the ligand perception of plant hormone co-receptors and provide scientific insight for developing novel molecular glues for regulating protein-protein interactions in living systems.

Limitations of the study
In the present study, we focused on the COI1-JAZ co-receptor in A. thaliana.However, previous studies have suggested differences in ligand perception by COI1-JAZ homologs in other plant species such as Solanum lycopersicum and Oryza sativa. 32,36,37The modes of action of (3R, 7S)-JA -JA-Ile and COR in other plant species were not identified in the present study.Therefore, it remains unclear whether the conclusions of this study apply equally to COI1-JAZ co-receptors in other plant species.In particular, monocotyledonous plants, including O. sativa, have multiple COI1 genes, and whether the modes of action of (3R, 7S)-JA-Ile and COR are the same requires further investigation.In addition, a (JASCO, Tokyo, Japan).All anhydrous solvents were either dried by standard techniques and freshly distilled before use or purchased in anhydrous form.Flash chromatography was performed on the Isolera system (Biotage Ltd., North Carolina, US).TLC was performed on Silica gel F254 (0.25 mm or 0.5 mm, MERCK, Germany).All reactions were carried out under air unless stated otherwise.SDS-PAGE was performed using Mini-Protean III (Bio-Rad Laboratories, Inc., US) devices.CBB-stained gel images were obtained using the Amersham Imager 680 detector (GE Healthcare, CA, US).Protein concentration was determined using NanoPhotometerN60 (IMPLEN GmbH, Germany).DSF experiments were carried out using StepOnePlus Real-Time PCR System (Thermo Fischer Scientific Inc., MA US) equipped with Protein Thermal Shift Software (Thermo Fischer Scientific Inc., MA US).ITC experiments were carried out using iTC 200 Microcalorimeter (MicroCal, Malvern Panalytical, UK) equipped with Origin (version 7.0) software (OriginLab Co., MA, USA).

Preparation of tag-free COI1 protein
The full-length Arabidopsis thaliana COI1 and ASK1 were co-expressed as a GST fusion protein in Sf9 insect cells. 19,25GST-COI1 and ASK1 P2 virus were infected Sf9 insect cells (final 1 3 10 6 cells/mL), which were cultured in 4 L of ESF921 medium matrix (2% FBS, 0.5% Penicillin-Streptomycin, 0.5 ng/mL Amphotericin B).After incubation for 72 h, the infected cells were enriched by centrifugation and subsequently lysed by sonication (Lysis buffer: 20 mM Tris-HCl, 200 mM NaCl, 10% Glycerol, 1 mM dithiothreitol, 1xcOmplete Protease inhibitor EDTA free, pH 8.0).Insoluble components and microsome fractions were removed by centrifugation (1000 g, 4 C, 30 min) and ultracentrifugation (150000 g, 4 C, 1.5 h), respectively.3.2 mL Glutathione Sepharose 4B (Cytiva, 80% resins in 4.0 mL EtOH, which was prewashed twice by 16 mL lysis buffer) was used to purify GST-COI1 protein in the cell lysate supernatant.The mixture was rotated for 1.5 h at 4 C and washed four times using a lysis buffer.The protein was solubilized in elution buffer (50 mM Tris-HCl, 100 mM NaCl, 10% Glycerol, 1 mM dithiothreitol, 0.1% Tween 20, 10 mM glutathione, pH 8.0).The first fraction was eluted by rotation for 20 min.This was followed by five times additional elution, and the supernatant was recovered by centrifugation.The obtained GST-COI1 fractions were concentrated into 2-3 mL by extra filtration using Amicon Ultra 15 (NMWL 10 kDa, Merck Millipore) (5000 g at 4 C).In iTC assay, this GST-COI1 solution was used with additional dialysis.For DSF assay, the solution of GST-COI1 was then transferred to Spectra/PorâDialysis Membrane (Spectrum Laboratories, Inc.) and was treated with TEV protease (New England BioLabs) in the 1 L dialysis buffer (20 mM Tris-HCl, 200 mM NaCl, 10% Glycerol, 1 mM dithiothreitol, pH 8.0) for ca. 12 h at 4 C to cleave the GST tag and to remove glutathione.The equilibrated resin and protein solutions were rotated for 1.5 h at 4 C to remove the GST components, leaving tag-free COI1 protein in the supernatant.The supernatant was recovered by centrifugation (1000 g, 4 C, 1 min).The resin was washed once, then the supernatant was recovered by centrifugation.

Preparation of tag-free ASK1 protein
The full-length Arabidopsis thaliana ASK1 was expressed as a His-tag-fusion protein (His-ASK1) in Sf9 insect cells.ASK1 P2 virus was infected Sf9 insect cells (final 1 3 10 6 cells/mL, MOI 1.5), which were cultured in 500 mL of PSFM-J1 medium matrix (2% FBS, 0.5% Penicillin-Streptomycin, 0.5 ng/mL Amphotericin B).After incubation for 72 h, the infected cells were enriched by centrifugation and subsequently lysed by sonication (Lysis buffer: 20 mM Tris-HCl, 200 mM NaCl, 10% Glycerol, 1 mM dithiothreitol, 1xcOmplete Protease inhibitor EDTA free, pH 8.0).Insoluble components and microsome fractions were removed by centrifugation (1000 g, 4 C, 30 min) and ultracentrifugation (150000 g, 4 C, 1.5 h), respectively.0.5 mL Ni Sepharose High Performance (Cytiva, prewashed twice by lysis buffer) was used to bind His-ASK1 protein in the cell lysate supernatant.The mixture was rotated for 1.5 h at 4 C and washed seven times using a lysis buffer.The tag-free ASK1 protein was solubilized by treating the resin with TEV protease (New England BioLabs) in the lysis buffer (20 mM Tris-HCl, 200 mM NaCl, 10% Glycerol, 1 mM dithiothreitol, pH 8.0) for ca.22 h at 4 C to cleave the His tag.

Differential scanning fluorometry (DSF) assay
COR/COR-MO/CFA-Ile were dissolved in DMSO, (3R,7S)/(3R,7R)-JA-Ile and cis-OPDA were dissolved in ethanol, JA was dissolved in sterilized water to generate 10 mM stock solutions and diluted with corresponding solvent for preparation of 100 mM and 1 mM stock solutions, respectively.JAZ1/9 peptides were dissolved in sterilized water to generate 10 mM stock solutions and diluted to prepare 100 mM stock solutions.The concentration of tag-free COI1 was determined by UV absorption at 280 nm using a calculated extinction coefficient value of 0.962.DSF experiments were carried out using StepOnePlus Real-Time PCR System (Thermo Fisher Scientific, Inc., MA, US).Protein Thermal Shift Dye (Thermo Fisher Scientific, Inc., MA, US) was used as the reporter dye.Reaction mixtures were prepared in PCR tubes, and each reaction was carried out on a 20 mL scale in HEPES buffer (50 mM HEPES-NaOH buffer, 100 mM NaCl, 20 mM 2-mercaptoethanol, 10% glycerol, 5 mM inositol-1,2,4,5,6-pentakisphosphate (IP5), pH 7.8) containing 1 mM tag-free COI1 protein, 3 mM JAZ peptide and designated concentration of ligand.Samples were heated from 25 C to 95 C. The denaturation curve was obtained using Protein Thermal Shift Software (Thermo Fisher Scientific, Inc., MA, US) to afford DTm value for each condition.An average DTm value of three independent measurements was plotted, and DTm dose-response curves were analyzed with nonlinear curve-fitting analysis to evaluate K a and K d values using KaleidaGraph software (v4.1.1,Synergy Software).

Isothermal titration calorimetry (iTC) assay
The experiments were performed with iTC 200 Microcalorimeter (MicroCal, Malvern Panalytical, UK) at 25 C. GST-COI1 concentration was determined by BCA assay (Thermo Fisher Scientific, Inc., MA, US) using NanoPhotometerN60 (IMPLEN GmbH, Germany).(3R, 7S)-JA-Ile (150 mM or 300 mM from 10 mM DMSO stock solution) in the reaction buffer (50 mM Tris, 100 mM NaCl, 10% Glycerol, 0.1% Tween 20, 10 mM IP5, pH 7.8) was loaded into the syringe and titrated against a mixture of JAZP1 only (10 mM from 1 mM stock solution) or GST-COI1 only (10 mM) or GST-COI1/JAZP1 (10 mM) in the reaction buffer in the cell.The measurements were carried out by 19 3 2 mL successive injections of solutions (initial injection of 0.4 mL), with a 180s spacing time between injections, stirring of 750 rpm, and a reference power of 5 mcal/s.The measured heat changes of the binding reactions were integrated and processed using the standard ''one set of sites'' model implemented in the Origin (version 7.0) software package to determine the binding stoichiometry (n) and the equilibrium dissociation constant (K d ) and other thermodynamic parameters (DG, DH, DS).Measurement between JA-Ile and GST-COI1 was analyzed with N = 1 manually due to the low C condition (C = n・M tot ・K a , n: stoichiometry, M tot : molar concentration of sample molecules in the cell, K a : association constant).

In-silico docking analysis
The structural preparation program MOE 2020.09(Chemical Computing Group) was used to deduce the structures of the absent residues in the crystal structure of the AtCOI1-JA-Ile-AtJAZ1 complex (PDB ID: 3OGL).The docking structure of AtCOI-CFA-Ile/JA-CMA-AtJAZ1 was prepared by replacing JA-Ile with CFA-Ile/JA-CMA in the docking simulation.A docking program in MOE was used for docking and Amber10:ETH force field parameters were assigned for the score estimations.

QUANTIFICATION AND STATISTICAL ANALYSIS
Curve fitting analysis of DTm dose-response curve The determination of the K d value was conducted by curve fitting the dose-response DTm values.Experiments were performed in triplicate to obtain the mean and S.D. (shown as error bars).Dose-response curves were analyzed with the nonlinear curve-fitting analysis to evaluate apparent K a and K d values using KaleidaGraph v4.1.1 software.

Figure 1 .
Figure 1.Perception model of JA ligands (A) Chemical structures of JA derivatives.(B) (3R, 7S)-JA-Ile acts as a molecular glue to induce protein-protein interaction between COI1 and JAZ.(C) COR is first perceived by COI1 and subsequently recruits JAZ to form a COI1-COR-JAZ ternary complex.

Figure 2 .
Figure 2. DSF assay of COR (A) Differential scanning fluorimetry (DSF) melting temperature curves of COI1 protein in the absence or presence of COR (0.3-30 mM).(B) DTm dose-response curves of COR.Experiments were performed in triplicate and data are shown as the mean G S.D. (C) Chemical structures of JAZP1/9 degron peptides.JAZ degron sequences are highlighted in yellow.(D) DSF melting temperature curves of COI1 protein with JAZP1/9 in the absence or presence of COR (0.3-30 mM).(E) DSF melting temperature curves of COI1 protein in the absence or presence of JAZP1/9 (1-30 mM).(F) DSF melting temperature curves of COI1 protein in the absence or presence of JA (0.3-30 mM).

Figure 3 .
Figure 3. DSF assay of COR-MO (A) Differential scanning fluorimetry (DSF) melting temperature curves of COI1 protein in the absence or presence of COR-MO (0.3-30 mM).(B) DTm dose-response curves of COR-MO.Experiments were performed in triplicate and data are shown as the mean G S.D. (C) DSF assays using COI1 and JAZP1/9 with COR-MO (0-30 mM).

Figure 6 .
Figure 6.DSF assay and docking analysis of CFA-Ile and JA-CMA (A) Chemical structures of CFA-Ile and JA-CMA.(B) Differential scanning fluorimetry (DSF) melting temperature curves of COI1 protein in the absence or the presence of CFA-Ile (0.3-100 mM).(C) DTm dose-response curves of CFA-Ile.Experiments were performed in triplicate and data are shown as the mean G S.D. (D) DSF melting temperature curves of COI1 protein in the absence or the presence of JA-CMA (6-600 mM).(3R, 7R)-JA-CMA: (3R, 7S)-JA-CMA = 95:5 mixture was used.(E) DTm dose-response curves of JA-CMA.Experiments were performed in triplicate and data are shown as the mean G S.D. (F) Docking structure of the COI1-JAZ1 crystal structure (PDB: 3OGM) with CFA-Ile.Docking simulations of CFA-Ile (cyan) were conducted against the ligand binding site of COI1 (gray) and JAZ1 in the crystal structure (PDB: 3OGM).Red dash bonds indicate the potential hydrogen bond.(G) Expanded view of the interface of CFA-Ile (cyan) and Phe89 (yellow) in the docking structure of COI1/CFA-Ile/JAZ1.(H) Docking structure of the COI1-JAZ1 crystal structure (PDB: 3OGM) with (3R, 7S)-JA-CMA.Docking simulations of (3R, 7S)-JA-CMA (green) were conducted against the ligand binding site of COI1 (gray) and JAZ1 in the crystal structure (3OGM).Red dash bonds indicate the potential hydrogen bond.