Abstract

Strigolactones (SLs) are multifunctional plant hormones regulating essential physiological processes affecting growth and development. In vascular plants, SLs are recognized by α/β hydrolase-fold proteins from the D14/DAD2 (Dwarf14/Decreased Apical Dominance 2) family in the initial step of the signaling pathway. We have previously discovered that N-phenylanthranilic acid derivatives (e.g. tolfenamic acid) are potent antagonists of SL receptors, prompting us to design quinazolinone and quinazolinedione derivatives (QADs and QADDs, respectively) as second-generation antagonists. Initial in silico docking studies suggested that these compounds would bind to DAD2, the petunia SL receptor, with higher affinity than the first-generation compounds. However, only one of the QADs/QADDs tested in in vitro assays acted as a competitive antagonist of SL receptors, with reduced affinity and potency compared with its N-phenylanthranilic acid ‘parent’. X-ray crystal structure analysis revealed that the binding mode of the active QADD inside DAD2's cavity was not that predicted in silico, highlighting a novel inhibition mechanism for SL receptors. Despite a ∼10-fold difference in potency in vitro, the QADD and tolfenamic acid had comparable activity in planta, suggesting that the QADD compensates for lower potency with increased bioavailability. Altogether, our results establish this QADD as a novel lead compound towards the development of potent and bioavailable antagonists of SL receptors.

Introduction

Strigolactones (SLs) are carotenoid-derived compounds that affect a large variety of developmental and physiological responses across the plant kingdom. In vascular plants, these compounds act as endogenous hormones to regulate processes such as shoot branching, root development, leaf senescence, and abiotic stress response [18]. SLs are also exuded from the roots of plants to the rhizosphere where they mediate interactions with symbiotic arbuscular mycorrhizal fungi [9] and with the Striga, Phelipanche, and Orobanche parasitic weeds [1013].

Details of SL biosynthesis, perception, and signaling are now well understood, particularly in model plants (see ref. [14] for a recent review), and α/β hydrolase-fold proteins from the Decreased Apical Dominance 2/Dwarf14 (DAD2/D14) clade have been identified as SL receptors in the first step of the SL signaling pathway [13,1519]. Unexpectedly for a receptor, these proteins still have low, but absolutely essential, hydrolase activity towards their hormone substrate that results in the cleavage of a bond between the ABC tricyclic lactone part of the SL molecule and the conserved butenolide moiety (D ring) [16,19]. During catalysis, a covalent intermediate is formed between the D ring and the histidine residue of the catalytic triad of the receptor, probably explaining the very slow rate of hydrolysis [15,18]. Binding and hydrolysis of the SL hormone trigger a large conformational change in the lid domain of the receptor, allowing its interaction with More Axillary Growth 2 (MAX2) protein from a Skp–Cullin–Fbox (SCF) protein complex [18,19]. SL receptors also show SL-dependent interactions with repressor proteins of SL signaling belonging to the D53/Suppressor of the MAX2-like (SMXL) family to probably recruit them to the SCF complex [17,20,21]. Following polyubiquitination by the SCF complex, the D53/SMXL repressor proteins are targeted to the 26S proteasome for degradation, resulting in a transcriptional response to the presence of the hormone signal.

SL receptors have a classical α/β hydrolase fold, consisting of a seven-stranded β-sheet ‘core’ domain flanked by seven α-helices. A lid, consisting of four additional helices inserted between β5 and α9 of the ‘core’ domain, caps the core domain over a large central cavity harboring the serine and histidine residues from the catalytic triad, as well as phenylalanine residues particularly conserved across members of the DAD2/D14 clade [16,19,2228]. These observations rapidly led to the hypothesis that SL receptors are ‘druggable’ and that the SL signaling pathway could therefore be manipulated by chemical factors to provide novel plant growth regulators affecting the plant architecture [29]. In line with this hypothesis, eight compound classes have recently been described as antagonists of SL receptors with various affinities, potencies, and in planta effects (reviewed in ref. [30]): soporidine [31], 2-methoxy-1-naphthaldehyde [32], β-lactones [33], DL1 [34], N-phenylanthranilic acid derivatives [35], Triton X-100 [36], carba-SLs [37], and 1,2,3-triazole ureas [38].

The mode of action of N-phenylanthranilic acid derivatives has been characterized at the structural level [35]. In the co-crystal structures of petunia DAD2 with tolfenamic acid and the related 2-(2′-methyl-3′-nitroanilino)benzoic acid (MNAB) and of rice D14 with MNAB, the antagonists were found to fully occupy the binding cavity with excellent shape complementarities. The binding mode of these compounds to the petunia and rice SL receptors is conserved, with only very minor variations seen in the respective conformations of the bound compounds in the rice vs. petunia receptor to account for the three amino acid differences observed in the binding cavities of these two proteins. Overall, N-phenylanthranilic acid derivatives bind through a combination of electrostatic and hydrophobic interactions. In particular, the carboxylic group of the antagonists anchors the compound deep inside the cavity through electrostatic interactions with the serine and histidine residues of the catalytic triad. Furthermore, an additional hydrogen bond is observed between the antagonists and Ser219 in DAD2 (Ser270 in rice D14). This residue, strictly conserved among members of the DAD2/D14 clade, sits at the tip of the so-called activation loop that defines the entrance of the internal cavity. In the crystal structure of D14 bound to D3 (the MAX2 orthologue in rice), the lid domain of the receptor has undergone large conformational changes to interact with D3 and the activation loop is completely disordered, suggesting that GR24-induced displacement of the activation loop may unlock the lid and activate the receptor. The interaction between N-phenylanthranilic acid derivatives and Ser219 supported this hypothesis where, by locking the activation loop in its ground (inactive) state, the antagonists prevent the opening of the lid and activation of the receptor.

A structure–activity relationship (SAR) study using 119 N-phenylanthranilic acid derivatives and 19 heteroanalog 2-phenoxybenzoic acid derivatives suggested that the core pharmacophore is N-(2-methylphenyl)anthranilic acid (Figure 1a), with possible substituents to be located on positions 3–5 of the Y ring [35] (ring labels are defined in Figure 1). The requirement for unsubstituted positions on the X ring, not unexpected given the tight steric constraints around that ring in the binding pocket, may also reflect a need to maximize internal hydrogen bonding by minimizing ring twist (Figure 1b). In the present study, we tested this hypothesis by designing a series of fused compounds related to 1-phenylquinazoline-2,4(1H,3H)-dione (Figure 1b,c) and assaying their ability to act as antagonists for SL receptors using biochemical, structural, and in planta methods.

Compound structures and synthetic scheme.

Figure 1.
Compound structures and synthetic scheme.

(a) N-(2-methylphenyl)anthranilic acid, the core pharmacophore as identified by the previous SAR study. (b) Schematic representation of ring closure hypothesis to maximize intramolecular hydrogen bonding and minimize ring twist, leading to 1-phenylquinazoline-2,4(1H,3H)-dione (0) as a novel reference compound. (c) 1–8: QADD and QAD structures mentioned and used in this study; 9: 2-(2′-methyl-3′-nitroanilino)benzoic acid (MNAB); 10: tolfenamic acid. Ring labels used throughout the manuscript are shown in red. (d) Synthetic scheme of QADDs.

Figure 1.
Compound structures and synthetic scheme.

(a) N-(2-methylphenyl)anthranilic acid, the core pharmacophore as identified by the previous SAR study. (b) Schematic representation of ring closure hypothesis to maximize intramolecular hydrogen bonding and minimize ring twist, leading to 1-phenylquinazoline-2,4(1H,3H)-dione (0) as a novel reference compound. (c) 1–8: QADD and QAD structures mentioned and used in this study; 9: 2-(2′-methyl-3′-nitroanilino)benzoic acid (MNAB); 10: tolfenamic acid. Ring labels used throughout the manuscript are shown in red. (d) Synthetic scheme of QADDs.

Experimental

Materials

Compounds 0, 5–8 were purchased from Molport (catalog nos: 003-809-289, 009-014-741, 002-817-647, 003-802-463, and 003-802-462, respectively), compound 10 was purchased from Sigma–Aldrich (catalog no. T0535), and compound 9 was obtained from William A. Denny (Auckland Cancer Society Research Centre).

Chemical syntheses

1-(3-Nitro-4-hydroxyphenyl)-1H-quinazoline-2,4-dione (2)

A solution of 1-(2-bromobenzoyl)-3-(3-nitrophenyl)-urea (250 mg, 0.68 mmol; the chemical synthesis of the intermediate urea is described in the Supplementary Experimental section) and freshly resublimed potassium tert-butoxide (450 mg) in dry N,N-dimethylformamide (DMF) (7 ml) was stirred at 75°C for 2.25 h. The mixture was cooled then poured onto aqueous HCl (0.5 M, 50 ml). The pale brown solid was filtered off, washed with water and then dried to give the crude product. Purification by column chromatography over silica gel eluting with dichloromethane to ethyl acetate (0–50%) gave the starting material, an unknown by-product and the title compound (9 mg). The identity of the title compound was shown by X-ray crystallography. HRMS-ESI [M−H] calcd for C14H8N3O5: 298.0469, Found: 298.0489. 1H and 13C NMR data are presented in Supplementary Table S1.

1-(3-Chloro-2-methylbenzyl)-1H-quinazoline-2,4-dione (3)

A solution of 1-(2-bromobenzoyl)-3-(3-chloro-2-methylphenyl)-urea (250 mg, 0.68 mmol; Supplementary Experimental section) and freshly resublimed potassium tert-butoxide (250 mg, 2 mmol) in dry DMF (5 ml) were heated to 70°C for 18 h. The mixture was cooled and then poured onto aqueous HCl (0.5 M, 50 ml). The pale brown solid was filtered off, washed with water and then dried to give the crude product. Purification by column chromatography over silica gel eluting with dichloromethane to ethyl acetate (0–50%) gave the pure product as a white solid (115 mg, 60%).

Microanalysis calcd for C15H11ClN2O2: C, 62.84; H, 3.87; N, 9.77. Found: C, 62.82; H, 3.99; N, 9.61. HRMS-ESI [M+Na]+ calcd for C15H11ClN2NaO2: 309.0401, found: 309.0384. 1H and 13C NMR data are presented in Supplementary Table S1.

1-(3-Chlorophenyl)-1H-quinazoline-2,4-dione (4)

A solution of 1-(2-bromobenzoyl)-3-(3-chlorophenyl)-urea (250 mg, 0.68 mmol; Supplementary Experimental section) and freshly resublimed potassium tert-butoxide (250 mg, 2 mmol) in dry DMF (5 ml) were heated to 70°C for 18 h. The mixture was cooled and then poured onto aqueous HCl (0.5 M, 50 ml). The pale brown solid was filtered off, washed with water, and then dried to give the crude product. Purification by column chromatography over silica gel eluting with dichloromethane to ethyl acetate (0–50%) gave the pure product as a white solid (165 mg, 85%).

Microanalysis calcd for C14H9ClN2O2: C, 61.67; H, 3.33; N, 10.27. Found: C, 61.57; H, 3.36; N, 10.16. HRMS-ESI [M+Na]+ calcd for C14H9ClN2NaO2: 295.0245, Found: 295.0223. 1H and 13C NMR data are presented in Supplementary Table S1.

Protein expression, purification, crystallization, and structure determination

Protocols for the expression and purification of wt-DAD2 (for in vitro assays) and DAD2Cys89Gln (for crystallization) have recently been described [35]. The Cys89Gln mutation was introduced to prevent intermolecular covalent bond formation between cysteine residues of different monomers and allowed the formation of diffracting crystals reproducibly. This mutation has been checked to have no effect on the in vitro activity of the protein [35]. Prior to crystallization, DAD2Cys89Gln was buffer exchanged into 20 mM Tris, pH 8.0, 150 mM NaCl using a Superdex 200 10/300 GL column (GE Healthcare) and concentrated to ∼6.5 mg/ml. DAD2Cys89Gln was incubated with QADD 2 (1.25 mM, i.e. ∼6× molar excess) in 20 mM Tris, pH 8.0, 150 mM NaCl, and 2% DMSO for 30 min at 18°C before crystallization. Drops consisted of 1 µl of protein solution and 1 µl of reservoir solution consisting of 0.1 M Tris–acetate, pH 8.5, 31% PEG 3350, and 0.2 M MgCl2. Crystals were cryo-protected by successive transfer to a reservoir solution containing 10%, 15%, and 20% glycerol. A full dataset was collected at the Australian Synchrotron MX1 beamline to 1.63 Å resolution. Data were processed with iMOSFLM [39] and pointless/aimless [40]. Structure determination was achieved using programs from the CCP4 package [41]. The structure was solved by molecular replacement using the DAD2 structure (PDB entry code: 4DNP) as starting models in Phaser [42]. Refinement cycles were carried out using Refmac5 [43] and Coot [44]. Restraint parameters for QADD 2 were obtained from JLigand [45]. Optimized refinement parameters obtained from the PDB_redo server [46] were used in the final stages of refinement. Data collection and refinement statistics are listed in Table 1.

Table 1
Data collection and refinement statistics
Crystal DAD2C89Q with QADD 2 
PDB entry code 6O5J 
Space group P
Cell parameters (Å and oa = 36.63, b = 56.83, c = 68.80 
α = 95.69, β = 94.59, ɣ = 108.74 
Dataset 
λ (Å) 0.9537 
 Reflections observed 243 675 
 Unique reflections 62 194 
 Resolution range (Å) 31.3–1.63 (1.66–1.63) 
 Rsym 0.08 (0.6) 
CC1/2 0.998 (0.757) 
 I/σ(I11.8 (2.2) 
 Completeness 96.0 (94.6) 
 Multiplicity 3.9 (3.9) 
Bwilson29.7 
Refinement 
 Resolution (Å) 29.37–1.63 (1.67–1.63) 
 Reflections 59 313 
 Total number of atoms 4842 
 TLS groups 
Rwork/Rfree 14.4/18.2 (20.3/21.6) 
 RMSD bonds (Å)/angles (o0.0156/1.94 
 Average B factors (Å2) (protein/water/inhibitor) 19.3/29.1/14.0 
 Ramachandran statistics of ϕ/ψ angles (%) Most favored: 90.5 
Additional favored: 9.1 
Generously favored: 0.4 
Disallowed: 0 
Crystal DAD2C89Q with QADD 2 
PDB entry code 6O5J 
Space group P
Cell parameters (Å and oa = 36.63, b = 56.83, c = 68.80 
α = 95.69, β = 94.59, ɣ = 108.74 
Dataset 
λ (Å) 0.9537 
 Reflections observed 243 675 
 Unique reflections 62 194 
 Resolution range (Å) 31.3–1.63 (1.66–1.63) 
 Rsym 0.08 (0.6) 
CC1/2 0.998 (0.757) 
 I/σ(I11.8 (2.2) 
 Completeness 96.0 (94.6) 
 Multiplicity 3.9 (3.9) 
Bwilson29.7 
Refinement 
 Resolution (Å) 29.37–1.63 (1.67–1.63) 
 Reflections 59 313 
 Total number of atoms 4842 
 TLS groups 
Rwork/Rfree 14.4/18.2 (20.3/21.6) 
 RMSD bonds (Å)/angles (o0.0156/1.94 
 Average B factors (Å2) (protein/water/inhibitor) 19.3/29.1/14.0 
 Ramachandran statistics of ϕ/ψ angles (%) Most favored: 90.5 
Additional favored: 9.1 
Generously favored: 0.4 
Disallowed: 0 

Values in parentheses are for the last resolution shell.

Docking

3D co-ordinates for compounds were generated in JLigand [45] from the CCP4 package, and pdb files were subsequently converted in pdbqt format using OpenBabel [47]. The DAD2 ‘empty’ structures (either the DAD2 structure bound to tolfenamic acid, PDB entry code 6AP6, or the DAD2 structure bound to QADD 2, where the bound compounds have been removed) were prepared in AutoDockTools [48] and used as receptors. Docking was performed in Autodock Vina [49] using a 25 × 25 × 25 Å3 box centered on the oxygen atom of Ser96. The maximum number of binding modes to be generated was set to nine.

Differential scanning fluorimetry

The differential scanning fluorimetry (DSF) experiments were performed as previously described [16]. DAD2 was first buffer exchanged into PBS using Superdex 75 10/300 GL (GE Healthcare). About 0.5 µl of compounds (10 mM in DMSO) were manually dispensed in a 384-well plate. In total, 18.5 µl of a solution containing 6.65 µM DAD2, SyproTangerine 10.25× in PBS were added to each compound using a BIOMEK 3000 pipetting robot (Beckman Coulter) to yield a final 19 µl reaction consisting of 6.48 µM DAD2, SyproTangerine 10×, 263 µM compound, and 2.6% DMSO in PBS. Reactions were incubated for 30 min at 18°C in the absence of light before DSF analysis.

Yeast two-hybrid assay

Yeast two-hybrid experiments were performed following methods from the Clontech Yeast Protocols Handbook, 2009, using yeast PJ69-4 [50], with DAD2 cloned into pBD vector [51], and MAX2A and D53A cloned into pAD vector [51]. The presence of the binding domain and activation domain fusion proteins in selected diploid yeast strains was confirmed by Western analysis [35]. Liquid culture assays using orthonitrophenyl-β-galactopyranoside (Sigma–Aldrich) as the substrate were used to quantify the strength of the interactions between DAD2 and target proteins. DAD2/MAX2A and DAD2/D53A interactions were detected in the presence of 5 and 1 µM GR24, respectively, and compared with DMSO controls. Each inhibitor was tested at 0.1, 1 and 10× molar ratios compared with GR24.

Intrinsic fluorescence assay

Experiments were performed on a FLUOStar Omega (BMG LabTech) using 280 ± 10 nm excitation and 340 ± 10 nm emission filters. The gain was set to 1800 and the number of flashes was set to 50. Compound stocks were prepared at 10× final concentration in 20 mM Tris (pH 8.0), 150 mM NaCl, and 20% DMSO. DAD2 was buffer exchanged in 20 mM Tris (pH 8.0), 150 mM NaCl using gel filtration (Superdex 75 10/300 GL, GE Healthcare) and its concentration was adjusted to 11.11 µM. For experiments, 10 µl of compounds were manually dispensed and then mixed with either 90 µl buffer or 90 µl DAD2 protein solution in flat-bottomed, black 96-well plates using a BIOMEK 3000 robot (Beckman Coulter). Final protein concentration was 10 µM, with compounds ranging from 0 to 200 µM in 20 mM Tris (pH 8.0), 150 mM NaCl, and 2% DMSO. All experiments (protein and buffer alike) were performed in triplicate. The plate was incubated for 30 min at 25°C before measurements. Binding curves were obtained by plotting the relative fluorescence (|ΔF|/F0) vs. compound concentration, where F0 is the fluorescence of the DMSO control, and |ΔF| = |F − F0|. GraphPad Prism was used to perform nonlinear regressions and determine the binding constants (Kd values) using the One site-Specific Binding model.

Enzymatic inhibition assays

Experiments were performed on a FLUOStar Omega (BMG LabTech) using 485 ± 12 nm excitation and 520 ± 10 nm emission filters. The temperature was set to 25°C and the gain was set to 920. Measurements were performed in black 96-well plates at 2-min intervals, with 20 flashes per cycle, over 90 min. Reactions were performed at seven Yoshimulactone Green (YLG) concentrations (0, 0.2, 0.4, 0.6, 1, 2, and 3.5 µM) and six inhibitor concentrations (0, 0.1, 0.5, 1, 5, and 10 µM). In all cases, the protein concentration was 0.34 µM and the reaction buffer was 20 mM Tris–HCl (pH 8.0), 150 mM NaCl, and 1% DMSO. All experiments were performed in triplicate. YLG-only controls were measured in the same buffer. Fluorescent units were converted to fluorescein concentrations using fluorescein standard curves. Michaelis–Menten analyses were performed by nonlinear regression in GraphPad Prism using the YLG series in the absence of inhibitor at 16 min. For inhibition kinetics, global nonlinear regression analyses were performed in GraphPad Prism using a competitive inhibition model.

Plant method

Arabidopsis bud assays were performed as previously described [16]. Sections of stems containing two buds were harvested from bolting plants. Treatments were applied to the base of the stems and contained 2% DMSO in 0.5× HSS [16]. Treatments comprised 0.5 µM GR24 and/or 5 µM tolfenamic acid and/or 5 µM QADD 2. After 5 days, bud length from both nodes was measured and normalized to the mock control.

Results and discussion

Initial docking study

We recently showed that docking of tolfenamic acid in a DAD2 ‘empty’ structure (i.e. the DAD2/tolfenamic acid crystal structure in which tolfenamic acid had been removed) almost perfectly matched the pose observed for this compound in the crystal structure, with a docking score of −10.6 kcal/mol [35]. Using the same methodology, we docked five commercially available QADD/QADs (0, 5–8, Figure 1c) and three proposed synthetic QADDs (2–4, Figure 1c) in DAD2's binding cavity. In all cases, the docking pose of the docked compound superimposed well with tolfenamic acid (Supplementary Figure S1). The corresponding docking scores ranged from −12.0 to −9.8 kcal/mol, suggesting that these compounds have a similar potential, if not better, to inhibit DAD2 than tolfenamic acid (Table 2).

Table 2
Docking scores of QADs/QADDs inside DAD2's binding cavity
Compound # Vina docking score (kcal/mol) 
0 −11.2 
2 −10.2 
3 −11.5 
4 −11.8 
5 −9.8 
6 −10.3 
7 −12.0 
8 −11.6 
Compound # Vina docking score (kcal/mol) 
0 −11.2 
2 −10.2 
3 −11.5 
4 −11.8 
5 −9.8 
6 −10.3 
7 −12.0 
8 −11.6 

Chemical syntheses

QADD 0 and QAD 5–8 are commercially available (see Experimental section). QADD 2–4 were synthesized using an adapted literature procedure [52]. The intermediate ureas were accessed through the treatment of 2-bromobenzamide with oxalyl chloride and the appropriate aniline (Supplementary Experimental section). The respective ureas were then dissolved in dry DMF and treated with freshly sublimed potassium tert-butoxide to afford the corresponding QADD, via an intramolecular nucleophilic substitution reaction (Figure 1d) [52]. The reaction proceeded smoothly to afford QADD 3 and 4. However, under these reaction conditions, the intramolecular cyclization of the urea precursor of QADD 1 gave an unexpected hydroxylation ortho to the nitro group to provide QADD 2 in low yield. The same molecule was subsequently made in higher yield by starting with 4-hydroxy-3-nitroaniline. QADD 2 was co-crystallized with DAD2 (see below) and the 1.63 Å resolution crystal structure confirmed the hydroxylated structure. While the mechanism of hydroxylation is unknown, there is a precedent for aromatic hydroxylation ortho to a nitro group, suspected to involve radical peroxide [53].

Binding and kinetics assays

To experimentally assess binding of QADs and QADDs to SL receptors, we tested whether QADD 0, 2–8 were able to trigger the stabilization of DAD2, AtD14, and OsD14 in a DSF assay. Using this assay, we recently showed that the thermal stability of the petunia, Arabidopsis, and rice SL receptors (DAD2, AtD14, and OsD14, respectively) is strongly increased upon the binding of inhibitors inside their internal cavity [35]. In an SAR study of 138 compounds closely related to tolfenamic acid, we further found that one compound, 2-(2′-methyl-3′-nitroanilino)benzoic acid (MNAB), triggered the strongest stabilization of all three receptors, with melting temperature shifts of +7.0, +4.4, and +3.5°C being observed for DAD2, AtD14, and OsD14, respectively, thereby providing a benchmark for the evaluation of the new QADDs 0, 2–8 [35]. As seen in Figure 2a, only QADD 2 triggered a stabilization of DAD2 although the thermal shift (ΔTm = +4.8°C) was lower than for MNAB. QADD 2 also stabilized AtD14 (ΔTm = +3.9°C, Figure 2b), but did not trigger thermal stabilization of OsD14 (ΔTm = −0.4°C, Figure 2c). In contrast with the in silico docking study that predicted all QAD/QADDs to efficaciously bind to DAD2, the other compounds showed no or only a small stabilizing effect on all three proteins (ΔTm < 1.6°C, Figure 2a–c). To confirm the DSF results, we measured the Kd values of QADD 2 to DAD2, AtD14, and OsD14 using intrinsic fluorescence (Figure 3a–c). Kd values of QADD 2 to DAD2 and AtD14 were 18.8 ± 1.8 (Figure 3a) and 19.2 ± 0.7 µM (Figure 3b), respectively, which represent ∼4× and ∼2× increases compared with the Kd values of MNAB to these two proteins [35]. In agreement with the DSF assay, the binding of QADD 2 to OsD14 was much weaker and the Kd value could not be determined accurately (Kd = 276 ± 119 µM, Figure 3c). Taken together, the DSF and intrinsic fluorescence assays showed that QADD 2 successfully binds to both DAD2 and AtD14, albeit with reduced affinity compared with MNAB; however, the binding of QADD 2 to OsD14 is very weak. Pre-steady-state inhibition kinetics using YLG, a profluorescent probe that releases fluorescein upon hydrolysis by SL receptors [13], have previously been used to estimate apparent Ki values of MNAB for DAD2, AtD14, and OsD14 (Ki = 0.16, 1.9, and 2.4 µM, respectively, [35]). Using the same setup, the apparent Ki values of QADD 2 for DAD2 and AtD14 were 1.03 ± 0.06 and 4.33 ± 0.32 µM, respectively, (Figure 3d,e) again showing that QADD 2 is less potent in inhibiting SL receptors in vitro than MNAB.

Thermal stability of SL receptors in the presence of QAD/QADD.

Figure 2.
Thermal stability of SL receptors in the presence of QAD/QADD.

DSF assay of DAD2 (a), AtD14 (b), and OsD14 (c) in the presence of QADs/QADDs. Top panels represent the experimental melting curves of each protein in the presence of QAD/QADD 0, 2–8 and in the presence of DMSO. Bottom panels are the derivatives of the melting curve from which melting temperatures are determined by the position of the minimum. Dotted lines indicate the melting temperature of proteins in the presence of DMSO (gray line) and in the presence of the best compound (colored lines). Melting temperature shifts are calculated according to: ΔTm = Tm (compound) − Tm (DMSO).

Figure 2.
Thermal stability of SL receptors in the presence of QAD/QADD.

DSF assay of DAD2 (a), AtD14 (b), and OsD14 (c) in the presence of QADs/QADDs. Top panels represent the experimental melting curves of each protein in the presence of QAD/QADD 0, 2–8 and in the presence of DMSO. Bottom panels are the derivatives of the melting curve from which melting temperatures are determined by the position of the minimum. Dotted lines indicate the melting temperature of proteins in the presence of DMSO (gray line) and in the presence of the best compound (colored lines). Melting temperature shifts are calculated according to: ΔTm = Tm (compound) − Tm (DMSO).

QADD 2 binding and inhibition.

Figure 3.
QADD 2 binding and inhibition.

(ac) Binding of QADD 2 to DAD2, AtD14, and OsD14, respectively, using intrinsic fluorescence. Each data point is the mean ± s.e.m. of three technical replicates. (d,e) YLG hydrolysis competition assays by QADD 2 for DAD2 and At14, respectively. Each data point is the mean ± s.e.m. of three technical replicates. For some data points, error bars are smaller than the symbol used for the data and do not appear on the graphs. All individual replicates for each compound concentration were included during the nonlinear global fit analysis using the competitive inhibition model.

Figure 3.
QADD 2 binding and inhibition.

(ac) Binding of QADD 2 to DAD2, AtD14, and OsD14, respectively, using intrinsic fluorescence. Each data point is the mean ± s.e.m. of three technical replicates. (d,e) YLG hydrolysis competition assays by QADD 2 for DAD2 and At14, respectively. Each data point is the mean ± s.e.m. of three technical replicates. For some data points, error bars are smaller than the symbol used for the data and do not appear on the graphs. All individual replicates for each compound concentration were included during the nonlinear global fit analysis using the competitive inhibition model.

We further showed that QADD 2 inhibits the GR24-dependent interaction between DAD2 and the downstream SL signaling targets PhMAX2A and PhD53A using yeast two-hybrid assays (Figure 4). However, in both cases, QADD 2 was ∼10× less effective than tolfenamic acid in these assays [35].

Yeast two-hybrid analysis.

Figure 4.
Yeast two-hybrid analysis.

Inhibition of (rac)-GR24-induced DAD2/PhMAX2A (a) and DAD2/PhD53A (b) interactions by QADD 2. Protein–protein interactions are quantified by assaying β-galactosidase activity in a yeast two-hybrid liquid culture system. Each data point is the mean ± s.e.m. of three technical replicates.

Figure 4.
Yeast two-hybrid analysis.

Inhibition of (rac)-GR24-induced DAD2/PhMAX2A (a) and DAD2/PhD53A (b) interactions by QADD 2. Protein–protein interactions are quantified by assaying β-galactosidase activity in a yeast two-hybrid liquid culture system. Each data point is the mean ± s.e.m. of three technical replicates.

Crystal structure of DAD2 in complex with QADD 2

Tolfenamic acid-related compounds inhibit SL receptors by binding inside their internal cavities through a combination of hydrophobic interactions with the side chains of four Phe residues lining the interior of the cavity (Phe27, Phe125, Phe158, and Phe194; all residue numbers refer to the DAD2 sequence) and electrostatic interactions with Ser96 and His246 from the catalytic triad, and with Ser219 from the activation loop [35]. Furthermore, the relative positions of the side chain of His246 and of the carboxylic group of tolfenamic acid-related compounds suggested that His246 could be protonated to form a salt bridge anchoring the bound compound inside the cavity [35]. To understand the discrepancies between the in silico docking studies and the experimental results, we co-crystallized DAD2 with QADD 2 and solved its structure to 1.63 Å resolution (Table 1). Excellent electron density was observed for QADD 2 inside DAD2's internal cavity (Figure 5) allowing detailed comparisons with the binding mode of MNAB, its closest parent compound (Figure 1). As seen in Figure 5, QADD 2 binds inside DAD2's cavity in a conformation totally different from that of MNAB: instead of having rings X and Y of both compounds superimposed, as predicted by the docking study, QADD 2 binds in a ‘back-to-front’ conformation where the X ring roughly occupies the position of the Y ring of MNAB, while the Y ring sits well above the X ring of MNAB and is rotated by ∼90° compared with the X ring of MNAB (Figure 6a). Consequently, while the docking study predicted that one of the carbonyl groups of the Z ring would interact with Ser96, it is in fact the serendipitous hydroxyl group on the Y ring that anchors QADD 2 to Ser96, while additional interactions occur between the nitro group of QADD 2, positioned deep inside the cavity, and the catalytic residues Ser96 and His246. In addition, one of the carbonyl groups on the Z ring forms a hydrogen bond with the hydroxyl group of Ser219 to stabilize the activation loop, thereby replacing the interaction previously seen between this residue and the carboxylic group of MNAB (Figure 6a). Despite the large differences observed between the binding mode and conformations of two compounds, only relatively small movements are observed for the protein residues lining DAD2's internal cavity (Figure 6a). Among these, rotations of the side chains of Phe125, Phe158, and Phe194 are observed to optimize interactions with rings Y and Z of QADD 2, and the side chain of Val193 is displaced by almost 2 Å to occupy some of the space previously filled by the X ring of MNAB (Figure 6a and Supplementary Figure S2). In contrast, the position of the side chain of Phe27, which was previously found to form an essential hydrophobic stacking interaction with the X ring MNAB [35] but now has only weak hydrophobic contacts with rings X and Y of QADD 2, remains unchanged. Likewise, the side chain of His218 has the same conformation as in the MNAB-bound structure, although this residue now has no direct contact with QADD 2 (Figure 6a). We next compared the binding site of QADD 2 with that of GR24 observed in the rice OsD14 structure [28]. As seen in Figure 6b, the two compounds overlap at the same position within the internal cavity, confirming that QADD 2 acts as a competitive inhibitor to prevent access to the active site cavity to the SL compound.

Electron density map of bound QADD 2.

Figure 5.
Electron density map of bound QADD 2.

The final sigmaA-weighted 2mFo − DFc electron density map contoured at 1.0σ around QADD 2 is shown in blue in two orientations (rotated ∼90° along the horizontal axis). QADD 2 and the catalytic serine (Ser96) are drawn in stick mode.

Figure 5.
Electron density map of bound QADD 2.

The final sigmaA-weighted 2mFo − DFc electron density map contoured at 1.0σ around QADD 2 is shown in blue in two orientations (rotated ∼90° along the horizontal axis). QADD 2 and the catalytic serine (Ser96) are drawn in stick mode.

QADD 2 acts as a competitive inhibitor.

Figure 6.
QADD 2 acts as a competitive inhibitor.

(a) Comparison of the binding modes of QADD 2 and MNAB inside DAD2's internal cavity. The DAD2/QADD 2 structure is drawn in green, while the DAD2/MNAB structure is drawn in pink. Hydrogen bonds are shown as dotted lines. (b) Comparison of the binding modes of QADD 2 and GR24 inside DAD2's internal cavity. The OsD14/GR24 structure (PDB 5DJ5, drawn in orange) was superimposed to the DAD2/QADD 2 structure in the same orientation as Figure 5a. OsD14 residues are labeled. QADD 2 (green) and GR24 (orange) are drawn in stick mode.

Figure 6.
QADD 2 acts as a competitive inhibitor.

(a) Comparison of the binding modes of QADD 2 and MNAB inside DAD2's internal cavity. The DAD2/QADD 2 structure is drawn in green, while the DAD2/MNAB structure is drawn in pink. Hydrogen bonds are shown as dotted lines. (b) Comparison of the binding modes of QADD 2 and GR24 inside DAD2's internal cavity. The OsD14/GR24 structure (PDB 5DJ5, drawn in orange) was superimposed to the DAD2/QADD 2 structure in the same orientation as Figure 5a. OsD14 residues are labeled. QADD 2 (green) and GR24 (orange) are drawn in stick mode.

Revisiting the docking study

The unexpected binding mode of QADD 2 inside DAD2's cavity prompted us to revisit the results of the docking study. To this end, the nine different docking poses obtained for QADD 2 in the DAD2/tolfenamic acid empty crystal structure (ranked in increasing order of their calculated binding affinity) were compared with the experimentally determined poses of both tolfenamic acid and of QADD 2 inside DAD2's cavity observed in the respective crystal structures (Supplementary Table S2A,B). Among these, docking poses #1 and, to a lesser extent, #5 have their ring structure superimposing well with tolfenamic acid and differ by a 180° rotation of their Y ring carrying the nitro group. Conversely, poses #4 and #6 share similarity with the conformation of QADD 2 seen in the crystal structure. Pose #6, in particular, shows a reasonably good agreement with the crystal structure, while pose #4 has its Y ring rotated almost 180° compared with the structure, therefore presenting the nitro group in the wrong direction.

To complement these results, we next performed the same docking experiment but using the empty DAD2/QADD 2 crystal structure as the receptor. As described earlier, only relatively small conformational differences are observed for a few residues lining DAD2's internal cavity when QADD 2 is bound instead of tolfenamic acid. However, these are sufficient to drive the docking results toward the correct solution, with docking pose #1 of QADD 2 now matching almost perfectly the conformation seen in the structure (Supplementary Table S2B). In addition, docking pose #3 is also very close to the conformation observed in the crystal structure, with only the Y ring differing by a 180° rotation. In contrast with the previous docking study, none of the docking poses obtained using the DAD2/QADD 2 empty structure as the receptor resembles the conformation seen for tolfenamic acid. This is most likely because of the position of Val193 in the DAD2/QADD 2 crystal structure that occupies some of the space previously filled by the X ring of tolfenamic acid (Figure 6). Overall, these results confirm that while docking can provide very valuable insights into the binding mode of small molecular compounds inside protein-binding pockets, it is also extremely sensitive to the local environment of the receptor. In the present case, the small conformational changes observed for a few of the residues lining the internal cavity of DAD2 upon binding QADD 2 could not have been anticipated or appropriately modeled to allow the initial docking study to accurately predict the back-to-front binding mode of QADD 2 inside the cavity. Although poses showing similarities to the true conformation of QADD 2 inside DAD2's cavity were present within the list of docking solutions (at #6 and, to a lesser extent, at #4), the top solution mimicking the conformation of tolfenamic acid/MNAB was the most plausible one given the knowledge available at the time the study was done.

Among all QADs/QADDs tested in this study, only QADD 2 harbors the hydroxyl group on the Y ring that directly interacts with the catalytic serine of the SL receptor at the bottom of the cavity, and none of the other QADs/QADDs were able to bind to the receptor. This confirms that anchoring of the compound through electrostatic interactions with the few polar residues present inside the binding cavity is essential for a successful interaction. A fundamental difference between N-phenylanthranilic acid derivatives and QAD/QADDs is the presence of a negatively charged carboxylic group in N-phenylanthranilic acid derivatives. In the crystal structure of DAD2 bound to tolfenamic acid, it was noted that the respective positions of the carboxylic group of tolfenamic acid and of the side chain of His246 from the catalytic triad suggested His246 to be protonated, thereby forming a salt bridge with the carboxylic group of the bound compound [35]. In contrast, the electronegative nitro group of QADD 2 does not point directly towards His246 but sits on the side and is further away. Furthermore, the lack of binding observed for QAD/QADD 6 (carrying a similar nitro group at the same position and lacking the additional hydroxyl group that interacts with Ser96) suggests that despite the electronegative character of the oxygen atoms, the presence of the nitro group is not sufficient to trigger a stable interaction with DAD2, and that interactions between the compound and both Ser96 and His246 are required for successful binding. Altogether, the lack of binding of most of the QADs/QADDs tested here and the lower binding affinity observed for QADD 2 indirectly support the hypothesis that a salt bridge between the carboxylic group of N-phenylanthranilic acid derivatives and the catalytic histidine of DAD2 deep inside the internal cavity is a key driver for the high affinity observed between DAD2 and this class of compounds.

In planta activity of QADD 2

In planta activity of QADD 2 was tested in Arabidopsis bud assays and compared with tolfenamic acid whose activity has been described recently [35]. As seen in Figure 7, QADD 2 showed effects on bud growth similar to tolfenamic acid both in the presence and absence of GR24. We noted recently that the relatively high concentration of tolfenamic acid required to give a physiological response in plants contrasts with the potency observed for this compound in vitro and suggested that uptake and/or transport of tolfenamic acid may be limiting factors for its activity in planta [35]. Indeed, intrinsic characteristics of tolfenamic acid suggest some potential drawbacks for in planta applications: a very low solubility in aqueous media and the presence of a charged group that may affect passive transport into plants. In contrast, we observe here that QADD 2, while less potent than tolfenamic acid in vitro, shows similar levels of activity when tested on bud growth. QADD 2 may therefore compensate a lower potency for the SL receptor by increased levels of solubility, uptake, and/or transport in planta, suggesting a fine balance between potency and bioavailability of candidate compounds to obtain successful plant growth regulators targeting the SL pathway. A firm conclusion on the relative efficacy of these two compounds will, however, require additional analyses. While further studies will also be required to fully understand the effects of different inhibitors on SL-regulated plant responses at both genetic and physiological levels, our results establish QADD 2 as a novel lead compound towards the generation of antagonists targeting plant SL receptors with improved potency and bioavailability.

In planta activity of QADD 2, compared with tolfenamic acid.

Figure 7.
In planta activity of QADD 2, compared with tolfenamic acid.

Arabidopsis bud growth assay treated with 0.5 µM GR24 and/or 5 µM compound (tolfenamic acid and QADD 2). Bud growth was normalized relative to the mock-treated control. The violin plots show the median (solid line) and quartiles values (dotted lines), and the area of each plot represents the frequency distribution of the data, n = 16. Statistical tests of differences between treatments were calculated by ANOVA and Fisher's protected LSD multiple comparisons test. Different letters indicate statistically significant results at P < 0.05. Data from the control and tolfenamic acid-treated buds have been published recently [35], and all data were collected at the same time.

Figure 7.
In planta activity of QADD 2, compared with tolfenamic acid.

Arabidopsis bud growth assay treated with 0.5 µM GR24 and/or 5 µM compound (tolfenamic acid and QADD 2). Bud growth was normalized relative to the mock-treated control. The violin plots show the median (solid line) and quartiles values (dotted lines), and the area of each plot represents the frequency distribution of the data, n = 16. Statistical tests of differences between treatments were calculated by ANOVA and Fisher's protected LSD multiple comparisons test. Different letters indicate statistically significant results at P < 0.05. Data from the control and tolfenamic acid-treated buds have been published recently [35], and all data were collected at the same time.

Database Depositions

The crystal structure of DAD2 in complex with 1-(3-nitro-4-hydroxyphenyl)-1H-quinazoline-2,4-dione (2) has been deposited in the Protein Data Bank, with entry code 6O5J.

Abbreviations

     
  • DMF

    N,N-Dimethylformamide

  •  
  • DSF

    differential scanning fluorimetry

  •  
  • MAX2

    More Axillary Growth 2

  •  
  • MNAB

    2-(2′-methyl-3′-nitroanilino)benzoic acid

  •  
  • QAD

    quinazolinone derivative

  •  
  • QADD

    quinazolinedione derivative

  •  
  • SAR

    structure–activity relationship

  •  
  • SCF

    Skp–Cullin–Fbox

  •  
  • SL

    strigolactone

  •  
  • YLG

    Yoshimulactone Green

Author Contribution

C.H., H.W.L., P.S., Z.L., and K.C.S. performed experiments. L.L., B.C.H., and N.B.P. synthesized the QADDs. C.H. wrote the manuscript with input from all authors. C.H. and K.C.S. conceived of and supervised the project.

Funding

This work was supported by funding from Plant and Food Research and a Marsden grant from the Royal Society Te Apārangi (contract no. PAF1301).

Acknowledgements

We thank William A. Denny from the Auckland Cancer Society for his analysis of our SAR study and initial suggestion of investigating QAD/QADDs, Revel S.M. Drummond for advice with the yeast two-hybrid experiments and members of the Laboratory of Structural Biology (University of Auckland) for help with X-ray data collection. This research was undertaken on the MX1 beamline at the Australian Synchrotron, part of ANSTO.

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Gomez-Roldan
,
V.
,
Fermas
,
S.
,
Brewer
,
P.B.
,
Puech-Pagès
,
V.
,
Dun
,
E.A.
,
Pillot
,
J.-P.
et al.  (
2008
)
Strigolactone inhibition of shoot branching
.
Nature
455
,
189
194
2
Kapulnik
,
Y.
,
Delaux
,
P.M.
,
Resnick
,
N.
,
Mayzlish-Gati
,
E.
,
Wininger
,
S.
,
Bhattacharya
,
C.
et al.  (
2011
)
Strigolactones affect lateral root formation and root-hair elongation in Arabidopsis
.
Planta
233
,
209
216
3
Pandey
,
A.
,
Sharma
,
M.
and
Pandey
,
G.K.
(
2016
)
Emerging roles of strigolactones in plant responses to stress and development
.
Front. Plant Sci.
7
,
434
4
Rasmussen
,
A.
,
Mason
,
M.G.
,
De Cuyper
,
C.
,
Brewer
,
P.B.
,
Herold
,
S.
,
Agusti
,
J.
et al.  (
2012
)
Strigolactones suppress adventitious rooting in Arabidopsis and pea
.
Plant Physiol.
158
,
1976
1987
5
Snowden
,
K.
,
Simkin
,
A.
,
Janssen
,
B.
,
Templeton
,
K.
,
Loucas
,
H.
,
Simons
,
J.
et al.  (
2005
)
The decreased apical dominance1/Petunia hybrida CAROTENOID CLEAVAGE DIOXYGENASE8 gene affects branch production and plays a role in leaf senescence, root growth, and flower development
.
Plant Cell
17
,
746
759
6
Umehara
,
M.
,
Hanada
,
A.
,
Yoshida
,
S.
,
Akiyama
,
K.
,
Arite
,
T.
,
Takeda-Kamiya
,
N.
et al.  (
2008
)
Inhibition of shoot branching by new terpenoid plant hormones
.
Nature
455
,
195
200
7
Woo
,
H.R.
,
Chung
,
K.M.
,
Park
,
J.-H.
,
Oh
,
S.A.
,
Ahn
,
T.
,
Hong
,
S.H.
et al.  (
2001
)
ORE9, an F-Box protein that regulates leaf senescence in Arabidopsis
.
Plant Cell
13
,
1779
1790
8
Yamada
,
Y.
,
Furusawa
,
S.
,
Nagasaka
,
S.
,
Shimomura
,
K.
,
Yamaguchi
,
S.
and
Umehara
,
M.
(
2014
)
Strigolactone signaling regulates rice leaf senescence in response to a phosphate deficiency
.
Planta
240
,
399
408
9
Akiyama
,
K.
,
Matsuzaki
,
K.
and
Hayashi
,
H.
(
2005
)
Plant sesquiterpenes induce hyphal branching in arbuscular mycorrhizal fungi
.
Nature
435
,
824
827
10
Al-Babili
,
S.
and
Bouwmeester
,
H.J.
(
2015
)
Strigolactones, a novel carotenoid-derived plant hormone
.
Annu. Rev. Plant Biol.
66
,
161
186
11
Brewer
,
P.B.
,
Koltai
,
H.
and
Beveridge
,
C.A.
(
2013
)
Diverse roles of strigolactones in plant development
.
Mol. Plant
6
,
18
28
12
Khosla
,
A.
and
Nelson
,
D.C.
(
2016
)
Strigolactones, super hormones in the fight against Striga
.
Curr. Opin. Plant Biol.
33
,
57
63
13
Tsuchiya
,
Y.
,
Yoshimura
,
M.
,
Sato
,
Y.
,
Kuwata
,
K.
,
Toh
,
S.
,
Holbrook-Smith
,
D.
et al.  (
2015
)
PARASITIC PLANTS. Probing strigolactone receptors in Striga hermonthica with fluorescence
.
Science
349
,
864
868
14
Lumba
,
S.
,
Holbrook-Smith
,
D.
and
McCourt
,
P.
(
2017
)
The perception of strigolactones in vascular plants
.
Nat. Chem. Biol.
13
,
599
606
15
de Saint Germain
,
A.
,
Clavé
,
G.
,
Badet-Denisot
,
M.A.
,
Pillot
,
J.P.
,
Cornu
,
D.
,
Le Caer
,
J.P.
et al.  (
2016
)
An histidine covalent receptor and butenolide complex mediates strigolactone perception
.
Nat. Chem. Biol.
12
,
787
794
16
Hamiaux
,
C.
,
Drummond
,
R.S.
,
Janssen
,
B.J.
,
Ledger
,
S.E.
,
Cooney
,
J.M.
,
Newcomb
,
R.D.
et al.  (
2012
)
DAD2 is an α/β hydrolase likely to be involved in the perception of the plant branching hormone, strigolactone
.
Curr. Biol.
22
,
2032
2036
17
Jiang
,
L.
,
Liu
,
X.
,
Xiong
,
G.
,
Liu
,
H.
,
Chen
,
F.
,
Wang
,
L.
et al.  (
2013
)
DWARF 53 acts as a repressor of strigolactone signalling in rice
.
Nature
504
,
401
405
18
Yao
,
R.
,
Ming
,
Z.
,
Yan
,
L.
,
Li
,
S.
,
Wang
,
F.
,
Ma
,
S.
et al.  (
2016
)
DWARF14 is a non-canonical hormone receptor for strigolactone
.
Nature
536
,
469
473
19
Zhao
,
L.H.
,
Zhou
,
X.E.
,
Wu
,
Z.S.
,
Yi
,
W.
,
Xu
,
Y.
,
Li
,
S.
et al.  (
2013
)
Crystal structures of two phytohormone signal-transducing α/β hydrolases: karrikin-signaling KAI2 and strigolactone-signaling DWARF14
.
Cell Res.
23
,
436
439
20
Stanga
,
J.P.
,
Morffy
,
N.
and
Nelson
,
D.C.
(
2016
)
Functional redundancy in the control of seedling growth by the karrikin signaling pathway
.
Planta
243
,
1397
1406
21
Zhou
,
F.
,
Lin
,
Q.
,
Zhu
,
L.
,
Ren
,
Y.
,
Zhou
,
K.
,
Shabek
,
N.
et al.  (
2013
)
D14-SCF(D3)-dependent degradation of D53 regulates strigolactone signalling
.
Nature
504
,
406
410
22
Bythell-Douglas
,
R.
,
Waters
,
M.T.
,
Scaffidi
,
A.
,
Flematti
,
G.R.
,
Smith
,
S.M.
and
Bond
,
C.S.
(
2013
)
The structure of the karrikin-insensitive protein (KAI2) in Arabidopsis thaliana
.
PLoS ONE
8
,
e54758
23
Guo
,
Y.
,
Zheng
,
Z.
,
La Clair
,
J.J.
,
Chory
,
J.
and
Noel
,
J.P.
(
2013
)
Smoke-derived karrikin perception by the α/β-hydrolase KAI2 from Arabidopsis
.
Proc. Natl Acad. Sci. U.S.A.
110
,
8284
8289
24
Kagiyama
,
M.
,
Hirano
,
Y.
,
Mori
,
T.
,
Kim
,
S.Y.
,
Kyozuka
,
J.
,
Seto
,
Y.
et al.  (
2013
)
Structures of D14 and D14L in the strigolactone and karrikin signaling pathways
.
Genes Cells
18
,
147
160
25
Nakamura
,
H.
,
Xue
,
Y.L.
,
Miyakawa
,
T.
,
Hou
,
F.
,
Qin
,
H.M.
,
Fukui
,
K.
et al.  (
2013
)
Molecular mechanism of strigolactone perception by DWARF14
.
Nat. Commun.
4
,
2613
26
Toh
,
S.
,
Holbrook-Smith
,
D.
,
Stogios
,
P.J.
,
Onopriyenko
,
O.
,
Lumba
,
S.
,
Tsuchiya
,
Y.
et al.  (
2015
)
Structure-function analysis identifies highly sensitive strigolactone receptors in Striga
.
Science
350
,
203
207
27
Xu
,
Y.
,
Miyakawa
,
T.
,
Nakamura
,
H.
,
Nakamura
,
A.
,
Imamura
,
Y.
,
Asami
,
T.
et al.  (
2016
)
Structural basis of unique ligand specificity of KAI2-like protein from parasitic weed Striga hermonthica
.
Sci. Rep.
6
,
31386
28
Zhao
,
L.H.
,
Zhou
,
X.E.
,
Yi
,
W.
,
Wu
,
Z.
,
Liu
,
Y.
,
Kang
,
Y.
et al.  (
2015
)
Destabilization of strigolactone receptor DWARF14 by binding of ligand and E3-ligase signaling effector DWARF3
.
Cell Res.
25
,
1219
1236
29
Nakamura
,
H.
and
Asami
,
T.
(
2014
)
Target sites for chemical regulation of strigolactone signaling
.
Front. Plant Sci.
5
,
623
30
Waters
,
M.T.
(
2019
)
Spoilt for choice: new options for inhibitors of strigolactone signaling
.
Mol. Plant
12
,
21
23
31
Holbrook-Smith
,
D.
,
Toh
,
S.
,
Tsuchiya
,
Y.
and
McCourt
,
P.
(
2016
)
Small-molecule antagonists of germination of the parasitic plant Striga hermonthica
.
Nat. Chem. Biol.
12
,
724
729
32
Mashita
,
O.
,
Koishihara
,
H.
,
Fukui
,
K.
,
Nakamura
,
H.
and
Asami
,
T.
(
2016
)
Discovery and identification of 2-methoxy-1-naphthaldehyde as a novel strigolactone-signaling inhibitor
.
J. Pest. Sci.
41
,
71
78
33
Xiang
,
H.
,
Yao
,
R.
,
Quan
,
T.
,
Wang
,
F.
,
Chen
,
L.
,
Du
,
X.
et al.  (
2017
)
Simple beta-lactones are potent irreversible antagonists for strigolactone receptors
.
Cell Res.
27
,
1525
1528
34
Yoshimura
,
M.
,
Sato
,
A.
,
Kuwata
,
K.
,
Inukai
,
Y.
,
Kinoshita
,
T.
,
Itami
,
K.
et al.  (
2018
)
Discovery of shoot branching regulator targeting strigolactone receptor DWARF14
.
ACS Central Sci.
4
,
230
234
35
Hamiaux
,
C.
,
Drummond
,
R.S.M.
,
Luo
,
Z.
,
Lee
,
H.W.
,
Sharma
,
P.
,
Janssen
,
B.J.
et al.  (
2018
)
Inhibition of strigolactone receptors by N-phenylanthranilic acid derivatives: structural and functional insights
.
J. Biol. Chem.
293
,
6530
6543
36
Shahul Hameed
,
U.
,
Haider
,
I.
,
Jamil
,
M.
,
Kountche
,
B.A.
,
Guo
,
X.
,
Zarban
,
R.A.
et al.  (
2018
)
Structural basis for specific inhibition of the highly sensitive ShHTL7 receptor
.
EMBO Rep.
19
,
e45619
37
Takeuchi
,
J.
,
Jiang
,
K.
,
Hirabayashi
,
K.
,
Imamura
,
Y.
,
Wu
,
Y.
,
Xu
,
Y.
et al.  (
2018
)
Rationally designed strigolactone analogs as antagonists of the D14 receptor
.
Plant Cell Physiol.
59
,
1545
1554
38
Nakamura
,
H.
,
Hirabayashi
,
K.
,
Miyakawa
,
T.
,
Kikuzato
,
K.
,
Hu
,
W.
,
Xu
,
Y.
et al.  (
2019
)
Triazole ureas covalently bind to strigolactone receptor and antagonize strigolactone responses
.
Mol. Plant
12
,
44
58
39
Battye
,
T.G.
,
Kontogiannis
,
L.
,
Johnson
,
O.
,
Powell
,
H.R.
and
Leslie
,
A.G.
(
2011
)
iMOSFLM: a new graphical interface for diffraction-image processing with MOSFLM
.
Acta Crystallogr. D Biol. Crystallogr.
67
(Pt
4
),
271
281
40
Evans
,
P.R.
and
Murshudov
,
G.N.
(
2013
)
How good are my data and what is the resolution?
Acta Crystallogr. D Biol. Crystallogr.
69
(Pt
7
),
1204
1214
41
Winn
,
M.D.
,
Ballard
,
C.C.
,
Cowtan
,
K.D.
,
Dodson
,
E.J.
,
Emsley
,
P.
,
Evans
,
P.R.
et al.  (
2011
)
Overview of the CCP4 suite and current developments
.
Acta Crystallogr. D Biol. Crystallogr.
67
(Pt
4
),
235
242
42
McCoy
,
A.J.
,
Grosse-Kunstleve
,
R.W.
,
Adams
,
P.D.
,
Winn
,
M.D.
,
Storoni
,
L.C.
and
Read
,
R.J.
(
2007
)
Phaser crystallographic software
.
J. Appl. Crystallogr.
40
(Pt
4
),
658
674
43
Murshudov
,
G.N.
,
Skubák
,
P.
,
Lebedev
,
A.A.
,
Pannu
,
N.S.
,
Steiner
,
R.A.
,
Nicholls
,
R.A.
et al.  (
2011
)
REFMAC5 for the refinement of macromolecular crystal structures
.
Acta Crystallogr. D Biol. Crystallogr.
67
(Pt
4
),
355
367
44
Emsley
,
P.
,
Lohkamp
,
B.
,
Scott
,
W.G.
and
Cowtan
,
K.
(
2010
)
Features and development of Coot
.
Acta Crystallogr. D Biol. Crystallogr.
66
(Pt
4
),
486
501
45
Lebedev
,
A.A.
,
Young
,
P.
,
Isupov
,
M.N.
,
Moroz
,
O.V.
,
Vagin
,
A.A.
and
Murshudov
,
G.N.
(
2012
)
JLigand: a graphical tool for the CCP4 template-restraint library
.
Acta Crystallogr. D Biol. Crystallogr.
68
(Pt
4
),
431
440
46
Joosten
,
R.P.
,
Joosten
,
K.
,
Cohen
,
S.X.
,
Vriend
,
G.
and
Perrakis
,
A.
(
2011
)
Automatic rebuilding and optimization of crystallographic structures in the Protein Data Bank
.
Bioinformatics
27
,
3392
3398
47
O'Boyle
,
N.M.
,
Banck
,
M.
,
James
,
C.A.
,
Morley
,
C.
,
Vandermeersch
,
T.
and
Hutchison
,
G.R.
(
2011
)
Open Babel: an open chemical toolbox
.
J. Cheminform.
3
,
33
48
Morris
,
G.M.
,
Huey
,
R.
,
Lindstrom
,
W.
,
Sanner
,
M.F.
,
Belew
,
R.K.
,
Goodsell
,
D.S.
et al.  (
2009
)
Autodock4 and AutoDockTools4: automated docking with selective receptor flexibility
.
J. Comput. Chem.
30
,
2785
2791
49
Trott
,
O.
and
Olson
,
A.J.
(
2010
)
Autodock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading
.
J. Comput. Chem.
31
,
455
461
50
James
,
P.
,
Halladay
,
J.
and
Craig
,
E.A.
(
1996
)
Genomic libraries and a host strain designed for highly efficient two-hybrid selection in yeast
.
Genetics
144
,
1425
1436
PMID:
[PubMed]
51
Maier
,
R.
,
Brandner
,
C.
,
Hintner
,
H.
,
Bauer
,
J.
and
Onder
,
K.
(
2008
)
Construction of a reading frame-independent yeast two-hybrid vector system for site-specific recombinational cloning and protein interaction screening
.
Biotechniques
45
,
235
244
52
Bowman
,
W.R.
,
Heaney
,
H.
and
Smith
,
P.H.G.
(
2003
)
Synthesis of 1H-quinazoline-4-ones using intramolecular aromatic nucleophilic substitution
.
Arkivoc
2003
,
434
442
53
Mattersteig
,
G.
,
Pritzkow
,
W.
and
Voerckel
,
V.
(
1990
)
The reaction of organic hydroperoxides with aromatic nitro compounds: first examples of a Vicarious Nucleophilic Hydroxylation
.
J. Prakt. Chem.
332
,
569
572

Author notes

*

Died 28 April 2018. We dedicate this paper to Lesley, in memory of her great enthusiasm and her relentless pursuit of synthetic targets.