Difference in the ligand affinity among redundant plant hormone receptors of rice OsCOI1a/1b/2-OsJAZs

ABSTRACT

(3R, 7S)-jasmonoyl-L-isoleucine (JA-Ile) is a lipid-derived plant hormone that regulates plant responses, including biotic/abiotic stress adaptation. In the plant cells, JA-Ile is perceived by COI1-JAZ co-receptor by causing protein-protein interaction between COI1 and JAZ proteins to trigger gene expressions. In this study, we focused on Oryza sativa, a model monocot and an important crop, with 45 possible OsCOI-OsJAZ co-receptor pairs composed of three OsCOI homologs (OsCOI1a, OsCOI1b, and OsCOI2) and 15 OsJAZ homologs. We performed fluorescein anisotropy and pull-down assays to examine the affinity between JA-Ile and OsCOI1a/1b/2-OsJAZ1-15 co-receptor pairs. The results revealed a remarkable difference in the modes of ligand perception by OsCOI1a/1b and OsCOI2. Recently, the unique function of OsCOI2 in some of the JA-responses were revealed. Our current results will lead to the possible development of OsCOI2-selective synthetic ligand.

Graphical Abstract

We demonstrated a remarkable difference in the modes of ligand perception among jasmonate co-receptor of Oryza sativa.

plant hormone, jasmonate, JA-Ile, OsCOI1-OsJAZ co-receptor, Oryza sativa

Land plants are sessile organisms and therefore have unique mechanisms to cope with many biotic and abiotic stresses they cannot escape, including pathogen infection, feeding by herbivorous insects, drought, and high salinity (Hirayama and Shinozaki 2010; Atkinson and Urwin 2012; Verma, Ravindran and Kumar 2016; Zhu 2016). Understanding and regulating molecular mechanisms of adaptation to these biotic/abiotic stresses will improve crop yield and quality. The lipid-derived plant hormone jasmonates (JAs) play a key role in regulating biotic/abiotic stress adaptation (Kazan 2015; Wasternack and Feussner 2018). In particular, the endogenous jasmonate (3R, 7S)-jasmonoyl-L-isoleucine (JA-Ile, Figure 1) functions as a primary plant hormone in the vascular plants. In Arabidopsis thaliana, a model dicot plant, JA-Ile is perceived by COI1-JAZ co-receptor by causing protein-protein interaction between the F-box protein CORONATINE INSENSITIVE 1 (COI1) and the transcriptional repressor protein JASMONATE-JIM DOMAIN (JAZ) (Chini et al.2007; Thines et al.2007; Howe, Major and Koo 2018). In this process, JA-Ile functions as a molecular glue. Subsequent degradation of the JAZ repressor through the 26S-proteasome mechanism derepresses transcription factors, leading to the expression of JA-responsive genes and activation of the JA signaling. The Arabidopsis COI1-JAZ co-receptor consists of one COI1 and one of the 13 JAZs encoded in the Arabidopsis genome. The resulting 13 COI1-JAZ co-receptor pairs are thought to function redundantly while also having unique functions in the regulation of JA signaling (Pauwels and Goossens 2011; Chini et al.2016). Novel and specific antagonists/agonists for the COI1-JAZ co-receptors have been developed as valuable chemical tools for studying jasmonate signaling (Monte et al.2014; Takaoka et al.2018; Chini et al.2021; Hayashi et al.2023).

Figure 1.

(a) Chemical structure of (3R, 7S)-JA-Ile (JA-Ile) and COR. (b) Pull-down assay using purified GST-OsCOI1 (5 nm) with FLAG-OsJAZ proteins in the presence of 0.1 µm of JA-Ile. HRP-conjugated anti-GST antibody (1/5000 dilution) was used to detect GST-COI1. Anti-FLAG antibody (1/5000 dilution) and HRP-conjugated anti-mouse IgG antibody (1/20 000 dilution) were used to detect FLAG-OsJAZ protein. A red arrowhead indicates each FLAG-OsJAZ protein. (c) K_d values of JA-Ile for possible OsCOI1-OsJAZ pairs obtained by fluorescence anisotropy assay. White denotes “not determined” due to the low increase of FA (Figure S7). Color represents the level of K_d value. The K_d value of JA-Ile was calculated by dose-response curves shown in Figure S8. (d) K_d values of COR for possible OsCOI1-OsJAZ pairs. White denotes “not determined” due to the low increase of FA (Figure S9). Color represents the level of K_d value. The K_d value of COR was calculated by dose-response curves shown in Figure S10.

In this paper, we focus on Oryza sativa, a model monocot and an important crop, which has a highly redundant COI-JAZ co-receptor system with 45 possible OsCOI-OsJAZ pairs composed of three OsCOI homologs (OsCOI1a, OsCOI1b, and OsCOI2) and 15 OsJAZ homologs (Figure S1) (Jain et al.2007; Mei, Zhou and Yang 2007; Ye et al.2009; Seo et al.2011; Lee et al.2013). Unlike Arabidopsis, O. sativa is characterized by the presence of multiple COI genes, and the functional diversity among the three OsCOI genes has attracted the attention of plant biologists. Phylogenetic analysis of the OsCOI genes revealed that OsCOI2 diverged due to gene duplication before OsCOI1a and OsCOI1b (Figure S1a), suggesting that OsCOI2 may have characteristic functions compared to the two OsCOI1 genes (Lee et al.2015). Recently, Inagaki et al. generated mutant lines of each OsCOI (Oscoi1a, Oscoi1b, and Oscoi2) and revealed that OsCOIs functioned redundantly in the JA-induced inhibition of growth (Inagaki et al.2022). They also demonstrated the crucial role of OsCOI2 in fertility and some of the JA-responses in O. sativa leaves. These results showed the redundancy and functional diversity of OsCOIs. Our comprehensive analyses of the affinity between JA-Ile and the 45 OsCOIs-OsJAZs co-receptor pairs showed a remarkable difference in the modes of ligand perception by OsCOI1a/1b and OsCOI2. In addition, in silico docking simulation demonstrated a remarkable difference in the modes of ligand perception by OsCOI1a/1b and OsCOI2. Our current results will lead to the possible development of an OsCOI2-selective synthetic ligand.

Materials and methods

Preparation of (3R, 7S)-JA-Ile

The diastereomeric mixture (147 mg, (3R,7S)/(3R,7R) = 5/95) was synthesized according to a previously reported method (Fonseca et al.2009) and separated by RP-HPLC as follows: 1^st separation was conducted using an ODS-MG-5 column (Develosil, Φ20 × 250 mm, eluent: 0.1:60:40 formic acid/H₂O/MeCN, flow rate: 8.0 mL/min, (3R,7R)-JA-Ile: Rt = 15.9 min, (3R,7S)- JA-Ile: Rt = 16.8 min) to afford (3R,7R)-JA-Ile (115 mg) as a white solid and slightly impure (3R,7S)- JA-Ile (7.62 mg) as a colorless oil. The latter fraction was further purified by ODS-MG-5 column (Develosil, Φ4.6 × 250 mm, eluent: 0.1:60:40 formic acid/H₂O/MeCN, flow rate: 1.0 mL/min, (3R,7R)-JA-Ile: Rt = 17.8 min, (3R,7S)-JA-Ile: Rt = 18.8 min) to afford pure (3R,7S)-JA-Ile (5.86 mg) as a colorless oil and (3R,7R)-JA-Ile (1.1 mg) as a white solid. Purified stereoisomers were analyzed by the 2^nd RP-HPLC condition (Figure S2).

Preparation of GST-fused OsCOIs

GST-fused OsCOI1a, OsCOI1b, and OsCOI2 were expressed in insect cells with AtASK1 and purified using Glutathione Sepharose 4B (GE Healthcare, Chicago, IL, USA) as previously described (Sheard et al.2010; Inagaki et al.2023).

JAZ protein expression using a cell-free expression system

FLAG-OsJAZ proteins were expressed using cell-free expression as previously described (Valea et al.2022).

Pull-down assay of OsCOI1-OsJAZ protein

0.1 µm JA-Ile and purified GST-OsCOI1 (5 nm) with ASK1 were dissolved in 350 µL of Tris buffer (50 mm Tris-HCl, pH 7.8, 100 mM NaCl, 10% Glycerol, 0.1% Tween20, 20 mm 2-mercaptoethanol (2-ME), and 1 µm inositol-1,2,4,5,6-pentakisphosphate, IP₅), and then FLAG-OsJAZs (10 µL of crude protein) was added. After incubation on ice for 1 h, the mixture was incubated with an anti-FLAG antibody (Sigma, F1804, 0.2 µL) at 4 °C for 10 h with rotation. After incubation, the mixture was combined with Protein G SureBeads (BIO-RAD, 10 µL in 50% Tris buffer slurry) and incubated at 4 °C for 1 h with rotation. After incubation, the sample was washed three times with PBS-T. The washed beads were resuspended in 35 µL of SDS-PAGE loading buffer containing dithiothreitol (DTT, 100 mm). After heating for 10 min at 60 °C, the samples were subjected to SDS-PAGE and analyzed by western blotting. SDS-PAGE and western blotting were performed using a Mini-Protean III electrophoresis apparatus (Bio-Rad, Hercules, CA, USA). Chemiluminescence was observed on Amersham Imager 680 (Cytiva, USA).

JAZ peptide synthesis

All JAZ peptides were prepared by microwave-assisted solid-phase synthesis with NovaSyn^® TGA resin 90 µm (OsJAZ11/15) or Fmoc-Tyr-wang resin (100-200 µm) using Initiator + Alstra (Biotage Ltd, NC, USA). The first Amino acid coupling reaction in OsJAZ11/15 was performed by adding N,N′-dicyclohexylcarbodiimide (DCC, 1.5 eq) and 4-dimethyl aminopyridine (DMAP, 1.5 eq) in DMF. After the peptide was fully elongated, the Fmoc-deprotected resin was mixed with 5-carboxy-fluorescein diacetate (3 eq), N,N,N′,N′-tetramethyl-O-(1H-benzotriazole-1-yl) uranium hexafluorophosphate (HBTU, 3 eq), 1-hydroxy benzotriazole monohydrate (HOBt·H₂O, 3 eq), and DIPEA (5 eq) in dry DMF at r.t. for 4 h. After the reaction, the peptide was deprotected by stirring with TFA: H₂O: TIS (95:2.5:2.5) at r.t. for 1.5 h (in cases of JAZ peptide that contains methionine, deprotection was performed with TFA: thioanisol: anisole: 1,2-ethanedithiol:: 90:5:2:3 for the avoidance of methionine oxidation). The eluted peptide was precipitated with Et₂O and collected by centrifugation (3500 g, 4 °C, 5 min) and purified by HPLC using a Develosil ODS-HG-5 column (Φ 4.6 × 250 mm) eluting with a linear gradient (CH₃CN (0.05% TFA): H₂O (0.05% TFA):: 20:80 (5 min) to 50:50 (35 min)) to collect fluorescein-conjugated JAZ peptide. After lyophilization, conjugated JAZ peptide was dissolved in sterilized water to prepare the stock solution (in cases of JAZ peptide that contains methionine, JAZ peptide was dissolved in sterilized water containing 1 mM TCEP to methionine oxidation). The concentration of these JAZ peptides was determined by their absorbance at 494 nm in an aqueous 0.1 N NaOH solution using a molar extinction coefficient of 75 000 M⁻¹cm⁻¹. The purity of these peptides was confirmed by HPLC analysis and characterized by MALDI-TOF MS as follows;

Fl-OsJAZ1P: m/z [M + H]⁺ calcd for 3678.94, found 3678.90

Fl-OsJAZ3P: m/z [M + H]⁺ calcd for 3415.79, found 3415.83

Fl-OsJAZ6P: m/z [M + H]⁺ calcd for 3593.88, found 3593.88

Fl-OsJAZ7P: m/z [M + H]⁺ calcd for 3612.85, found 3612.85

Fl-OsJAZ8P: m/z [M + H]⁺ calcd for 3581.80, found 3581.80

Fl-OsJAZ9P: m/z [M + H]⁺ calcd for 3478.81, found 3478.80

Fl-OsJAZ10P: m/z [M + H]⁺ calcd for 3569.78, found 3569.78

Fl-OsJAZ11P: m/z [M + H]⁺ calcd for 3564.92, found 3564.94

Fl-OsJAZ12P: m/z [M + H]⁺ calcd for 3743.91, found 3743.91

Fl-OsJAZ13P: m/z [M + H]⁺ calcd for 3600.86, found 3600.85

Fl-OsJAZ14P: m/z [M + H]⁺ calcd for 3757.04, found 3757.06

Fl-OsJAZ15P: m/z [M + H]⁺ calcd for 2784.51, found 2784.45.

Ultraviolet (UV)-visible spectra were recorded on a UV-2600 spectrophotometer (Shimadzu, Kyoto, Japan). MALDI-TOF MS analyses were performed on a 4800 plus MALDI TOF/TOF Analyzer (AB Sciex, Framingham, MA, USA).

Fluorescence anisotropy measurement

Fluorescence anisotropy titration experiments were performed at 25 °C in 50 µL of Tris buffer (50 mm Tris-HCl, pH 7.8, 100 mm NaCl, 10% Glycerol, 0.1% Tween20, 20 mm 2-mercaptoethanol (2-ME), and 1 µm inositol-1,2,4,5,6-pentakisphosphate, IP₅) containing 10 nmOsCOI1 and 10 nm Fl-OsJAZPs in a 96-well clear-bottom plate unless otherwise noted. Each ligand was added to the solution and incubated at 25 °C until the r-value did not fluctuate, and anisotropy intensities were recorded (λ_ex/λ_em = 485 nm/516 nm). R values were calculated using the following equation: r = (I_VV—G × I_VH)/(I_VV + 2 G × I_VH). I_VV and I_VH are the fluorescence intensities observed through polarizers parallel and perpendicular to the polarization of the exciting light, respectively. G is a correction factor to account for instrumental differences in detecting emitted compounds. Fluorescence anisotropy was recorded on an EnVision 2105 (PerkinElmer, USA).

Pull-down assay of OsCOI1-OsJAZ peptide

Each ligand and purified GST-OsCOI1 (5 nm) with ASK1 was dissolved in 350 µL of Tris buffer, and then Fl-OsJAZPs (10 nm) was added. After incubation on ice for 1 h, the mixture was combined with an anti-Fluorescein antibody (Abcam, ab19491, 0.2 µL) and incubated at 4 °C for 5 h with rotation. After incubation, the mixture was combined with Protein A Mag Sepharose Xtra (GE Healthcare, 10 µL in 50% Tris buffer slurry) and incubated at 4 °C for 1 h with rotation. After incubation, the sample was washed five times with PBS-T. The washed beads were resuspended in 35 µL of SDS-PAGE loading buffer containing dithiothreitol (DTT, 100 mm). After heating for 10 min at 60 °C, the samples were subjected to SDS-PAGE and analyzed by western blotting. SDS-PAGE and western blotting were performed using a Mini-Protean III electrophoresis apparatus (Bio-Rad, Hercules, CA, USA). Chemiluminescence was observed on Amersham Imager 680 (Cytiva, USA).

In silico analysis

The homology modeling of OsCOI1 was obtained based on the crystal structure of the AtCOI1-JA-Ile-AtJAZ1 complex (PDB ID: 3OGL). The structural preparation program MOE 2020.09 was used to deduce the structures of the absent residues in AtCOI1. OsJAZ3 was constructed by mutation of the residues in AtJAZ1. Each snapshot was obtained after energy minimization in the MOE program. The structure of OsCOIs-CFA-Phe-OsJAZ3 was prepared by replacing COR with CFA-Phe 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.

Sequence and phylogenetic analyses

The amino acid sequence of OsCOI1s and OsJAZs were obtained from The Rice Annotation Project Database (RAP-db; https://rapdb.dna.affrc.go.jp/). The amino acid sequence of ZmCOI and TaCOI was obtained from Genbank (https://www.ncbi.nlm.nih.gov/genbank/) and The International Wheat Genome Sequencing Consortium (IWGSC; https://www.wheatgenome.org/), respectively. Phylogenetic analysis for aligning the full-length protein sequences was performed with Geneious (Tomy Digital Biology Co., Ltd.). The Pam250 program was used for sequence alignment, and the Neighbor-Joining program (Saitou and Nei 1987) was used for drawing the result trees.

Results and discussion

COI1-JAZ-dependent bioactivity of various jasmonates has been examined by genetic and biochemical analyses. On Arabidopsis COI1-JAZ co-receptor, biochemical studies on the affinity of jasmonates with COI1-JAZ were performed using pull-down, yeast two-hybrid (Y2H), radioactive ligand binding, and fluorescence anisotropy assays (Fonseca 2009; Sheard 2010; Takaoka et al.2019; Hu et al.2022), In this study, we performed pull-down and fluorescein anisotropy (FA) assays to examine the affinity between JA-Ile and OsCOI1a/1b/2-OsJAZ1-15 co-receptor pairs (Figure 1b and c). We prepared GST-fused OsCOI1a/1b/2 proteins (GST-OsCOI1a/1b/2) in an insect-cultured cell expression system (Sheard 2010; Li et al.2017) and full-length OsJAZ proteins by a cell-free expression system (Valea 2022). In this assay, we used a naturally occurring JA-Ile as a ligand (Figure 1). We prepared pure JA-Ile of (3R, 7S)-stereochemistry as a ligand: a mixture of (3R, 7S)-JA-Ile: (3R, 7R)-JA-Ile (5:95) prepared by a previous method (Miyamoto et al.2019) was subsequently separated using chiral HPLC to afford the pure JA-Ile (Figure S2). The result of the pull-down assay showed that JA-Ile (0.1 mm) bound to OsCOI1a-OsJAZ3/4/6 and OsCOI1b-OsJAZ3/4/6/7 co-receptor pairs and to most combinations of OsCOI2-OsJAZs (OsCOI2-OsJAZ1-13/15; Figure 1b and Figure S3). Additionally, quantitative analysis of the affinity between JA-Ile and GST-OsCOI1a/1b/2-OsJAZ1-15 pairs was performed using an FA assay (Takaoka 2019). It was performed using fluorescein-labeled OsJAZ1-15 short peptides (OsJAZPs) instead of full-length OsJAZ proteins. OsJAZPs contain the ligand-interacting domain of full-length OsJAZs (Figure S4-S6) (Inagaki 2022). We observed no affinity with OsCOI1a/1b-OsJAZP2/5 pairs, a weak affinity with OsCOI1a/1b-OsJAZP1/6-15 pairs and a OsCOI2-OsJAZ2/5, and moderate to strong affinity with OsCOI1a/1b-OsJAZP3/4 and OsCOI2-OsJAZP1/3/4/6-15 pairs (Figure 1c and Figures S7 and S8). In addition, the same trend was observed using coronatine (COR), a structural and biological mimic of JA-Ile, (Xie et al.1998) in the FA assay (Figure 1d and Figures S9 and S10). Upon comparing the binding affinity (K_d) between JA-Ile/COR and each OsCOI-OsJAZP pair, the affinity of JA-Ile/COR with OsCOI2-OsJAZP pairs was relatively higher than that with OsCOI1a/1b-OsJAZP pairs. Thus, a significant difference was observed between the ligand affinities of OsCOI1a/1b-OsJAZPs and OsCOI2-OsJAZPs.

We posited that the cause of JA-Ile's binding to most of OsCOI2-OsJAZ pairs was the dissimilarity in size and shape of the ligand-binding pockets of OsCOI1a/1b and OsCOI2. To examine the ligand binding modes of OsCOI1a/1b and OsCOI2, we performed in silico docking simulations using OsCOI1a/1b/2-COR-OsJAZ3 homology models generated from the crystal structure of Arabidopsis COI1-JAZ1 in complex with JA-Ile (PDB: 3OGL) (Figure 2a-c) (Sheard 2010). We focused on the amino acid residues that contact JA-Ile within the ligand binding pocket of OsCOI1a/1b/2 homology models (Figure 2d). Owing to the mutation from ⁹⁴Tyr/³⁸⁹Tyr in OsCOI1a and ⁹⁶Tyr/³⁹¹Tyr in OsCOI1b to ⁹¹Phe/³⁹¹His in OsCOI2, OsCOI2 has a larger ligand-binding pocket than OsCOI1a/1b (Figure 2c). Thus, it is expected that JA-Ile in the binding pocket of OsCOI2 can take flexible conformation appropriate for the binding to the OsJAZs (OsJAZ1-15) of variable conformations. Thus, unlike OsCOI1a/1b, OsCOI2 was expected to accommodate a larger ligand than JA-Ile, and a ligand larger than JA-Ile could have OsCOI2-selectivity. Plants synthesize JA-L-amino acid conjugates (JA-AAs) other than JA-Ile (Yan et al.2016; Fu et al.2022). Some JA-AAs, including JA-Ile, JA-Val, JA-Leu, JA-Met, and JA-Phe, are known to accumulate in O. sativa after herbivorous attacks (Xiao et al.2014; Fu 2022). JA-AAs with large amino acid side chain are possible candidates of OsCOI2-selective ligand.

Figure 2.

(a-c) In silico homology models of the ligand binding pockets OsCOI1s-JA-Ile-OsJAZ3. These homology modeling of (a) OsCOI1a, (b) OsCOI1b, and (c) OsCOI2 were generated by using AtCOI1-JA-Ile-AtJAZ1 crystal structure (PDB: 3OGL) as a template. Yellow dotted lines mean hydrogen bonds. JA-Ile was shown as a green-colored stick model. The binding pocket of OsCOI1a-OsJAZ3, OsCOI1b-OsJAZ3, and OsCOI2-OsJAZ3 are represented with a grey-colored surface. Electron densities of ⁹⁴Tyr/³⁸⁹Tyr, ⁹⁶Tyr/³⁹¹Tyr, and ⁹¹Phe/³⁹¹His are represented with magenta-colored surfaces. (d) Alignment comparison of ligand-contacting amino acid residues between AtCOI1 and OsCOI1a/COI1b/COI2.

Recent research has provided insights into the JA signaling in O. sativa (Liu et al.2015; Okada, Abe and Arimura 2015; Trang Nguyen et al.2019). In particular, it has been demonstrated that individual components of the OsCOIs-OsJAZs system, such as OsCOI1b or OsJAZ1/8, play distinct roles in the genetically redundant co-receptor system (Yamada et al.2012; Lee 2015; Liu 2015; Wu et al.2015; Tian et al.2019; Trang Nguyen 2019; Feng et al.2020). Based on the phylogeny of three OsCOI proteins, the amino acid sequences between OsCOI1a/1b and OsCOI2 are quite different: The amino acid sequence identity between OsCOI1a and OsCOI1b is 82%, while the identity between OsCOI1a/1b and OsCOI2 is 63% (Figure S11). And it has been reported that COR first binds to COI1 and then recruits JAZ to form a COI1-JAZ co-receptor (Yan et al.2018), suggesting that the three OsCOI homologs have different affinities for the ligand and may have different functions.

The difference in ligand perception between OsCOI1a/OsCOI1b and OsCOI2 was also suggested by Li et al. (Lee 2013). They prepared the Arabidopsis mutant lines in which the OsCOI1a or OsCOI1b gene was overexpressed in the Arabidopsis coi1-1 background, and demonstrated that these two mutant lines could restore the JA signaling and successfully complement the function of the COI1 gene in A. thaliana. In contrast, overexpression of the OsCOI2 gene could not complement the coi1-1 mutant in A. thaliana. These results also suggest that endogenous JA-Ile may not be responsible for the interaction between OsCOI2 and AtJAZs. Therefore, it is believed that JA-Ile interacts with OsCOI2 in a distinct binding mode compared to its interaction with AtCOI1, potentially leading to the induction of other JA signaling pathways in O. sativa.

Recently, the development of Oscoi1a/Oscoi1b/Oscoi2 mutant lines has revealed the difference and redundancy in the function of OsCOI1a/1b and OsCOI2 (Inagaki 2022; Wang et al.2023). Three OsCOIs functioned redundantly in JA-induced growth inhibition, and OsCOI2 plays a crucial role in some of JA-dependent responses, including fertility (Inagaki 2023). Therefore, currently found difference of ligand affinity between OsCOI1a/1b and OsCOI2 will provide a clue for further development of a synthetic ligand that selectively binds to OsCOI2-OsJAZs co-receptors which will be a valuable chemical tool for deciphering and regulating jasmonate signaling in O. sativa.

Our result demonstrated that distinct difference in affinity was observed between JA-Ile and OsCOI1a/OsCOI1b/OsCOI2-OsJAZs co-receptors (Figure 1). And in silico docking studies using homology models (Figure 2) suggested that OsCOI2 has a larger ligand-binding pocket than OsCOI1a/1b. Unlike representative dicots such as A. thaliana and tomato, which have a single COI1 gene in their genomes, monocots have genetically redundant COI genes, including OsCOI1a/1b/2 in O. sativa, ZmCOI1a/1b/1c/1d/2a/2b in Zea mays, and TaCOI1A/2/3/4/5/6/7/8A in wheat (Triticum aestivum L.) (Wang et al.2005; An et al.2018; Bai et al.2018; Qi et al.2022). Phylogenetic analyses have shown that ZmCOI2b in Z. mays, and TaCOI5a in T. aestivum are orthologs of OsCOI2, which belong to the same subgroup 12 of the COI family gene (Figure S12a) (Bai 2018). Furthermore, we aligned the amino acid residues, which are expected to be in contact with the ligand in the ligand-binding pocket of ZmCOIs and TaCOIs. We compared them between subfamily 12 COIs (ZmCOI2b and TaCOI5a) and other COIs in the same plant (Figure S12b). The results showed that ZmCOI2b and TaCOI5a have a smaller amino acid residue (³⁹³H in ZmCOI2b and ³⁷⁵H in TaCOI5a) than other COIs in Z. may and T. aestivum. These findings suggest that a larger synthetic ligand than JA-Ile, such as JA-Phe, JA-Tyr, and JA-Trp, may have selective affinity among the highly redundant COIs in other monocots and could be valuable chemical tools for further studies in the function of highly redundant COI genes in monocots.

Acknowledgments

We thank for Dr. Kengo Hayashi (Ritsumeikan University) for his technical assistances.

Data Availability

All relevant data are contained in the manuscript.

Author contribution

Conceptualization, M.U.; methodology, M.U., T.O., T.K., and T.K.; validation, M.U., and T.K.; formal analysis, T.O., T.K., T.K., and M.U.; investigation, T.O., T.K., T.K. H.U., and K.M.; resources, T.K. H.U., K.M., H.I., K.M., and K.O.; data curation, M. U. and T.K.; writing—original draft preparation, M.U.; writing—review and editing, M.U., T.O. T.K., and T.K., K.M., and T.O.; visualization, T.O. and T.K.; supervision, M.U.; project administration, M.U.; funding acquisition, M.U. All authors have read and agreed to the published version of the manuscript.

Funding

This work was financially supported by a Grant-in-Aid for Scientific Research from JSPS, Japan (nos. 23H00316, 23H04883, 22KK0076, 21K19037, 20H00402, JPJSBP120229905, and JPJSBP120239903 for MU), and Nagase Science and Technology Foundation (MU).

Disclosure statement

The authors declare no conflict of interest.

Author notes

These authors contributed equally.