In the first committed step of histidine biosynthesis, adenosine 5′-triphosphate (ATP) and 5-phosphoribosyl-α1-pyrophosphate (PRPP), in the presence of ATP phosphoribosyltransferase (ATP-PRT, EC 2.4.2.17), yield phosphoribosyl-ATP. ATP-PRTs are subject to feedback inhibition by histidine that allosterically binds between the regulatory domains. Histidine biosynthetic pathways of bacteria, lower eukaryotes, and plants are considered promising targets for the design of antibiotics, antifungal agents, and herbicides because higher organisms are histidine heterotrophs. Plant ATP-PRTs are similar to one of the two types of their bacterial counterparts, the long-type ATP-PRTs. A biochemical and structural study of ATP-PRT from the model legume plant, Medicago truncatula (MedtrATP-PRT1) is reported herein. Two crystal structures, presenting homohexameric MedtrATP-PRT1 in its relaxed (R-) and histidine-bound, tense (T-) states allowed to observe key features of the enzyme and provided the first structural insights into an ATP-PRT from a eukaryotic organism. In particular, they show pronounced conformational reorganizations during R-state to T-state transition that involves substantial movements of domains. This rearrangement requires a trans- to cis- switch of a peptide backbone within the hinge region of MedtrATP-PRT1. A C-terminal α-helix, absent in bacteria, reinforces the hinge that is constituted by two peptide strands. As a result, conformations of the R- and T-states are significantly different from the corresponding states of prokaryotic enzymes with known 3-D structures. Finally, adenosine 5′-monophosphate (AMP) bound at the active site is consistent with a competitive (and synergistic with histidine) nature of AMP inhibition.

Introduction

Prokaryotes, lower eukaryotes, and plants are histidine autotrophs in contrast to histidine-heterotrophic animals. This makes histidine biosynthetic pathways attractive and promising targets in the design of antibiotics, antifungal agents, and herbicides [13]. In plants, histidine is biosynthesized in a route resembling that of bacteria [4,5]. The plant pathway involves eight enzymes, numbered consecutively, disregarding here their actual functions: HisN1 [6]; HisN2 [7], HisN3 [8], HisN4 [9], HisN5 [6], HisN6 [10], HisN7 [6,11], and HisN8 [1214] that altogether catalyze eleven reactions (HisN2, HisN4, and HisN8 are bifunctional). Unlike for the bacterial orthologs, structural information about the histidine biosynthesizing enzymes of plant origin is limited, as only structures of HisN5 [15,16], HisN7 [17], and HisN8 [18] were reported.

In the first committed step of histidine biosynthesis, HisN1, which is an adenosine 5′-triphosphate (ATP) phosphoribosyltransferase (ATP-PRT, EC 2.4.2.17), utilizes ATP and 5-phosphoribosyl-α1-pyrophosphate (PRPP) to yield phosphoribosyl-ATP (PR-ATP, Figure 1A), with the release of inorganic pyrophosphate (PPi) as a byproduct [19,20]. ATP-PRTs require Mg2+ for the activity. The reaction equilibrium is shifted towards the substrates unless PPi is removed from the milieu or decomposed, e.g. by an inorganic pyrophosphatase [20]. It has been also shown that the substrate binding is sequential, with ATP binding first and PR-ATP leaving last [21] and that no covalent intermediate is formed with the enzyme over the course of the reaction [22,23].

ATP-PRTs utilize ATP and 5-phosphoribosyl-α1-pyrophosphate (PRPP) to yield phosphoribosyl-ATP (PR-ATP).

Figure 1.
ATP-PRTs utilize ATP and 5-phosphoribosyl-α1-pyrophosphate (PRPP) to yield phosphoribosyl-ATP (PR-ATP).

A scheme of the chemical reaction catalyzed by ATP-PRTs is shown in (A). (B) Sequence alignment of selected ATP-PRT representatives: Medicago truncatula MedtrATP-PRT1 [UniProt accession number: G7JFL4]; Arabidopsis thaliana AtATP-PRT1 [Q9S762]; AtATP-PRT2 [Q8GSJ1]; Escherichia coli EcATP-PRT [P60757]; Campylobacter jejuni CjATP-PRT [Q9PM78]; Mycobacterium tuberculosis MtATP-PRT [P9WMN1]. Numbering above the sequences and the secondary structure elements (α helices, yellow pipes; β strands, blue arrows) correspond to MedtrATP-PRT1. (C) Kinetic analysis of MedtrATP-PRT1 with ATP and PRPP. The measurements were fit to a non-linear curve in Prism version 6.07 software (GraphPad), based on the Michaelis–Menten equation, to calculate apparent KM and kcat. The corresponding second substrate concentrations were 2 mM ATP and 1 mM PRPP. Error bars, S.D. calculated from four independent replications. (D) Inhibition by L-histidine in the absence and presence of AMP (three experiments); 2 mM ATP and 1 mM PRPP were used.

Figure 1.
ATP-PRTs utilize ATP and 5-phosphoribosyl-α1-pyrophosphate (PRPP) to yield phosphoribosyl-ATP (PR-ATP).

A scheme of the chemical reaction catalyzed by ATP-PRTs is shown in (A). (B) Sequence alignment of selected ATP-PRT representatives: Medicago truncatula MedtrATP-PRT1 [UniProt accession number: G7JFL4]; Arabidopsis thaliana AtATP-PRT1 [Q9S762]; AtATP-PRT2 [Q8GSJ1]; Escherichia coli EcATP-PRT [P60757]; Campylobacter jejuni CjATP-PRT [Q9PM78]; Mycobacterium tuberculosis MtATP-PRT [P9WMN1]. Numbering above the sequences and the secondary structure elements (α helices, yellow pipes; β strands, blue arrows) correspond to MedtrATP-PRT1. (C) Kinetic analysis of MedtrATP-PRT1 with ATP and PRPP. The measurements were fit to a non-linear curve in Prism version 6.07 software (GraphPad), based on the Michaelis–Menten equation, to calculate apparent KM and kcat. The corresponding second substrate concentrations were 2 mM ATP and 1 mM PRPP. Error bars, S.D. calculated from four independent replications. (D) Inhibition by L-histidine in the absence and presence of AMP (three experiments); 2 mM ATP and 1 mM PRPP were used.

Bacterial species possess either of the two kinds of ATP-PRTs, the short- or the long-type forms [24]. Both types utilize a conserved catalytic mechanism [25] and are subject to allosteric feedback inhibition by histidine, similar to other biosynthetic pathways, wherein endproducts control activities of entry enzymes. The short ATP-PRTs form heterooctamers with HisZ, a histidyl-tRNA synthetase paralog responsible for the feedback regulation [26]. Plant ATP-PRTs belong to the long-type subfamily (Figure 1B), as plant genomes seem to lack sequences coding for short-type ATP-PRTs. Initially, the bacterial long-type ATP-PRTs were thought to exist in a histidine-controlled equilibrium between active, dimeric and inactive, hexameric species [27,28]. However, later they were shown to strictly form homohexamers and that the allosteric binding of histidine impacts shape of the homohexameric assembly instead of the oligomeric state per se [2931]. More precisely, feedback regulation of ATP-PRT activity was linked to twisting of the regulatory domain [2729]. Although plant ATP-PRTs from Arabidopsis thaliana (At), AtATP-PRT1, and AtATP-PRT2 were also shown to be feedback-inhibited by histidine [32], the feature was missing a structural rationale.

Prokaryotic long-type APT-PRTs are composed of three domains (I, II, and III), of which the first two form a catalytic core, whereas the third one serves a regulatory function [2729]. The six catalytic sites are formed at crevices between domains I and II and are exposed to the void in the center of the protein assembly. ATP-PRTs are competitively inhibited by adenosine 5′-monophosphate (AMP) (and ADP), i.e. when the ATP/AMP ratio is low the pathway does not consume ATP. Moreover, the two inhibitors, histidine and AMP, show a synergistic effect [33]. The need to post-translationally control the histidine production reflects a high metabolic cost of the biosynthesis; the complete pathway requires over thirty ATP molecules to produce each histidine molecule [34].

Bacterial enzymes involved in the histidine biosynthesis have been extensively studied and still are one of the major interests of antibiotic design. This project is aimed at completing the landscape of histidine biosynthesis in plants to provide a scaffold for novel herbicides in view of the increasing glyphosate resistance. Pinpointing structural similarities and differences between bacterial and plant enzymes appears necessary to reach a high specificity of ATP-PRT-targeted agents against bacteria and plants. In addition, discovering the structural elements that stand for the feedback inhibition in plants may lead to production of crops in which this feature is mitigated. With this in mind, a thorough structural study of Medicago truncatula ATP-PRT (MedtrATP-PRT1) is presented herein. The MedtrATP-PRT1 acronym is used to avoid confusion with the enzyme from Mycobacterium tuberculosis (MtATP-PRT), which is frequently referred to in this article. This is the first structure of an ATP-PRT enzyme from a eukaryotic organism. The structure was solved experimentally using the single anomalous dispersion (SAD) method. This work is focused on a comparison with bacterial ATP-PRTs with known three-dimensional structures. Moreover, structural changes that are caused by histidine binding which transforms the enzyme from relaxed (R-) to tense (T-) state are discussed. Finally, AMP binding site, which agrees with the AMP competitive inhibition towards both substrates is shown.

Materials and methods

Cloning, overexpression, and purification of MedtrATP-PRT1

The total RNA was isolated from M. truncatula leaves using the RNeasy Plant Mini Kit (Qiagen) and was reverse-transcribed into the complementary DNA (cDNA). The chloroplast-targeting peptide was recognized using the TargetP 1.1 server [35,36], and the produced construct was designed to yield MedtrATP-PRT1 N-truncated at Thr25. The open reading frame coding for MedtrATP-PRT1 25–373 fragment was amplified by polymerase chain reaction (Forward: TACTTCCAATCCAATGCCACTCATCATCAAGTTCTTAATGGGAACACA, Reverse: TTATCCACTTCCAATGTTACTAAAGCCCAAGTTTAGAAAGGAGCTGA). The expression plasmid, based on pMCSG68 vector (Midwest Center for Structural Genomics), was created by the ligase-independent cloning method [37]. The insert correctness was confirmed by DNA sequencing.

BL21 Gold E. coli cells (Agilent Technologies) served as hosts for overexpression in LB media supplemented with 150 μg/ml ampicillin. The cultures were incubated with shaking at 190 rpm at 37°C until the A600 reached 1.0. Then, they were chilled to 18°C, and isopropyl-D-thiogalactopyranoside was added at a final concentration of 0.5 mM to start overexpression carried out for 18 h. The cell pellet from the 2 l culture was centrifuged at 3500×g for 30 min at 4°C and resuspended in 35 ml of binding buffer [50 mM Tris–HCl pH 8.0; 500 mM NaCl; 20 mM imidazole; 1 mM tris(2-carboxyethyl)phosphine (TCEP)] and stored at −80°C. The selenomethionine (SeMet)-labeled protein was obtained using the same E. coli clone by inhibition of methionine biosynthesis (addition of inhibitory amino acids) and supplementation with SeMet [38].

The cells were disrupted by sonication in an ice/water bath using 4-s bursts and 26-s cooling intervals for a total of 5 min of the probe working time. The cell debris was spun down by centrifugation at 25 000×g for 30 min at 4°C. The supernatant was mixed with 3 ml of HisTrap HP resin (GE Healthcare) and poured into a 50 ml column plugged into vacuum pump-VacMan setup (Promega). The resin-bound MedtrATP-PRT1 was washed six times, each with 40 ml of the binding buffer, and eluted with 20 ml of elution buffer (50 mM Tris–HCl pH 8.0; 500 mM NaCl; 400 mM imidazole; 1 mM TCEP). The His6-tag was cleaved with TEV protease (at final concentration 0.1 mg/ml) overnight and the imidazole concentration was lowered to 20 mM by simultaneous dialysis at 4°C. The sample was mixed with fresh HisTrap resin, and the flow-through (containing MedtrATP-PRT1) was collected in which the cleaved His6-tag and the His6-tagged TEV protease had been eliminated. The sample was concentrated to 2.4 ml and applied on a HiLoad Superdex 200 16/60 column (GE Healthcare), equilibrated with a buffer composed of 25 mM Tris–HCl pH 8.0, 150 mM NaCl, and 1 mM TCEP. The final, purified protein contained Ser-Asn-Ala linker, preceding the genuine Thr25 of MedtrATP-PRT1. The SeMet-labeled protein was purified analogously.

Crystallization and diffraction data collection

The hexameric fraction of MedtrATP-PRT1 was concentrated using centrifugal concentrators (Millipore) to 17 mg/ml (based on A280 with the extinction coefficient of 28 400 per subunit). The crystals were grown by vapor diffusion method in hanging drops containing 2 µl the protein and 2 µl of the reservoir solutions. MedtrATP-PRT1 R-state crystals (both native and SeMet-labeled) grew in 2.63 M NaCl, 100 mM Bis-Tris propane pH 6.3, and were cryoprotected by a brief wash in the solution supplemented with 20% glycerol. To obtain the T-state complex with histidine and AMP, the ligands (5 mM each) were added to the protein solution supplemented with 10 mM MgCl2 and 100 mM KCl and incubated for 1 h. The reservoir solution was 2.53 M NaCl, 100 mM Bis-Tris propane pH 6.3, and 10% glycerol which allowed for a direct crystal vitrification in liquid nitrogen. Data were collected at 22-ID and 19-ID beamlines at the Advanced Photon Source, Argonne, U.S.A. The diffraction images for the SeMet-labeled crystals were processed with HKL3000 [39], whereas XDS [40] was used for the other two. The statistics of the data collection and processing are summarized in Table 1.

Table 1
Data collection and refinement statistics.

Values in parentheses correspond to the highest resolution shell.

MedtrATP-PRT1: R-state T-state SeMet-labeled 
Data collection 
Beamline APS 22-ID APS 19-ID APS 22-ID 
Wavelength (Å) 0.9780 0.9793 0.9778 
Space group P21 P21 P21 
Unit cell parameters 
  a, b, c (Å) 98.8, 193.5, 99.1 89.2, 178.7, 92.1 99.7, 195.9, 100.0 
  β (°) 95.9 105.7 95.9 
Resolution (Å) 50.0–2.92 (3.09–2.92) 50.0–2.88 (2.95–2.88) 99.0–3.80 (3.87–3.80) 
Unique reflections 82 460 (12 719) 62 118 (4580) 37 381 (1838) 
Multiplicity 4.2 (3.9) 3.8 (3.9) 9.4 (9.0) 
Completeness (%) 99.1 (95.3) 99.1 (99.5) 99.9 (100.0) 
 Rmeas1(%) 7.6 (105.2) 5.5 (92.6) 11.0 (29.8) 
<I/σ(I)15.2 (1.8) 18.4 (1.8) 20.8 (9.0) 
Refinement 
 Rfree reflections 1237 1056  
No. of atoms (non-H) 
  protein 15 480 15 181  
  ligands 5 × glycerol, 6 × Cl 6 × His, 6 × AMP  
  solvent 40 20  
 Rwork/Rfree (%) 19.7/25.0 19.2/23.9  
Average B-factor (Å2
  protein 102 108 (85)2  
  Ligands 116 105 (80)2  
  solvent 60 75  
rmsd from ideal geometry 
  bond lengths (Å) 0.008 0.007  
  bond angles (o1.3 1.0  
Ramachandran statistics (%) 
  favored 95.0 96.6  
  allowed 4.4 3.4  
  outliers 0.6 0.0  
PDB code 6czl 6czm  
MedtrATP-PRT1: R-state T-state SeMet-labeled 
Data collection 
Beamline APS 22-ID APS 19-ID APS 22-ID 
Wavelength (Å) 0.9780 0.9793 0.9778 
Space group P21 P21 P21 
Unit cell parameters 
  a, b, c (Å) 98.8, 193.5, 99.1 89.2, 178.7, 92.1 99.7, 195.9, 100.0 
  β (°) 95.9 105.7 95.9 
Resolution (Å) 50.0–2.92 (3.09–2.92) 50.0–2.88 (2.95–2.88) 99.0–3.80 (3.87–3.80) 
Unique reflections 82 460 (12 719) 62 118 (4580) 37 381 (1838) 
Multiplicity 4.2 (3.9) 3.8 (3.9) 9.4 (9.0) 
Completeness (%) 99.1 (95.3) 99.1 (99.5) 99.9 (100.0) 
 Rmeas1(%) 7.6 (105.2) 5.5 (92.6) 11.0 (29.8) 
<I/σ(I)15.2 (1.8) 18.4 (1.8) 20.8 (9.0) 
Refinement 
 Rfree reflections 1237 1056  
No. of atoms (non-H) 
  protein 15 480 15 181  
  ligands 5 × glycerol, 6 × Cl 6 × His, 6 × AMP  
  solvent 40 20  
 Rwork/Rfree (%) 19.7/25.0 19.2/23.9  
Average B-factor (Å2
  protein 102 108 (85)2  
  Ligands 116 105 (80)2  
  solvent 60 75  
rmsd from ideal geometry 
  bond lengths (Å) 0.008 0.007  
  bond angles (o1.3 1.0  
Ramachandran statistics (%) 
  favored 95.0 96.6  
  allowed 4.4 3.4  
  outliers 0.6 0.0  
PDB code 6czl 6czm  
1

Rmeas = redundancy independent R-factor [72].

2

Value for subunit A.

Determination and refinement of the crystal structures

The crystal structure of R-state MedtrATP-PRT1 was solved by SAD using SeMet-labeled protein. For phasing, data from two crystals were merged. The phasing was performed with HKL3000 [39] which utilizes SHELXC/D/E [41]. The initial model was built using the native data (R-state) with Phenix.AutoBuild [42], and was placed inside the unit cell with the ACHESYM server [43]. COOT [44] was used for manual fitting in the electron density maps between rounds of model refinement in Refmac [45] (R-state) with TLS [46] groups. Individual domains of the R-state model served to solve the T-state structure by molecular replacement with PHASER [47]. The T-state structure, in which the chains B and F are not well defined in the electron density maps, was refined with non-crystallographic symmetry (NCS) and secondary structure restraints [48] in Phenix.refine [49]. The refinement statistics are listed in Table 1.

Enzyme kinetics assay

Steady-state measurements were performed with Biotek Powerwave XS2 at 24°C using the method developed by Ames et al. [20], which follows PR-ATP formation (ε290 = 3600 M−1 cm−1). E. coli inorganic pyrophosphatase (EcPPase, at 4 µM, hexamer) was used to drive the reaction towards PR-ATP. MedtrATP-PRT1 (100 nM, hexamer) was preincubated in a buffer containing 50 mM Tris–HCl pH 8.0, 10 mM MgCl2, 100 mM KCl, 50 mM NaCl, 1 mM TCEP with the appropriate ATP concentration for 5 min and the reactions were initiated by the addition of PRPP. The assay was performed at pH 8.0 to reflect the milieu of chloroplast stroma in daylight [50], whereas Mg2+ concentration was increased to 10 mM to ensure optimal turnover of EcPPase. Both substrates, ATP and PRPP were tested at 0–4 mM concentrations. Saturating concentrations used were 2 mM ATP and 1 mM PRPP (also during inhibition studies). Histidine as an inhibitor was assayed at 0–400 µM, and at 0–200 µM when 2 mM AMP was included. AMP did not show inhibitory properties up to 8 mM. Experiments were performed in four replicates (KM, kcat) or three replicates (inhibition studies). Data were analyzed using nonlinear regression curve fitting in Prism 6.07 (GraphPad).

Other software used

Molecular illustrations were created with UCSF Chimera [51], which also served for calculations of root-mean-square-deviations (rmsds) for Cα atom pairs within 3 Å radius. Secondary structure elements were assigned using PDBsum [52]. The surface conservation was analyzed with ConSurf [53] based on 500 sequences from Uniprot [54] that sampled a list of 1520 unique HMMER [55] homologues with identities between 25 and 100%. Surface electrostatic potential distribution was calculated using PDB2PQR and APBS servers [56,57]. Because parts of chains B and F are disordered in the T-state structure, figures showing oligomers of the T-state are created by copying the chain ‘A' by NCS operators. Sequence identities/similarities were calculated in BLAST [58]. Signal peptides were predicted with TargetP [35] webserver.

Results and discussion

Overall properties of MedtrATP-PRT1

Two isoforms of chloroplastic A. thaliana ATP-PRTs (AtATP-PRT1 and AtATP-PRT2; UniProt IDs: Q9S762 and Q8GSJ1, respectively) were previously characterized from a functional perspective [32], but a structural study of a plant enzyme has been missing. Based on a BLAST search [59], Medicago truncatula also contains two ATP-PRT isoforms (UniProt IDs: MedtrATP-PRT1, G7JFL4; and MedtrATP-PRT2, G7INM4). Disregarding the plastid-targeting peptide at the N-termini, both sequences are nearly identical, with the exception of a 22-residue insert in MedtrATP-PRT2 that maps in between Gly116 and Gln117 of MedtrATP-PRT1. Because this insert is not conserved in any other protein, it possibly is an artifact of in silico splicing prediction. Therefore, as a more reliable subject, MedtrATP-PRT1 was used in this study. MedtrATP-PRT1 shares 82.6 and 80.3% sequence identity with AtATP-PRT1 and AtATP-PRT2, respectively (Figure 1B). Plant ATP-PRTs (excluding target peptides) are over 340 residues-long (Figure 1B). On the contrary, sequences of bacterial ATP-PRTs are significantly shorter: Escherichia coli and Campylobacter jejuni ATP-PRTs (EcATP-PRT and CjATP-PRT) consist of 299 residues each (UniProt ID: P60757 and Q9PM78, respectively), whereas MtATP-PRT has only 284 residues (P9WMN1).

Enzymatic tests verified that the obtained MedtrATP-PRT1 is an active enzyme (Figure 1C). The apparent Michaelis–Menten constants for ATP (470 µM, Table 2) and PRPP (182 µM) are higher than those generally observed for bacterial orthologs [29,30], but close to the values reported for AtATP-PRT1 (600 and 130 µM, respectively) and AtATP-PRT2 (510 and 570 µM) [32]. MedtrATP-PRT1 is inhibited by histidine (IC50 = 30.1 µM, Table 2, Figure 1D), similar to other ATP-PRTs. Moreover, AMP at 2 mM lowered the value nearly five-fold (Table 2, Figure 1D); notably, in this setup ATP and AMP were at 1:1 ratio. Interestingly, AMP alone did not show inhibitory properties up to 8 mM, unlike in the case of, e.g. CjATP-PRT (IC50 (AMP) = 2.1 mM) [60].

Table 2
Kinetic constants of MedtrATP-PRT1.
Parameter MedtrATP-PRT1 
KM ATP (µM) 470 ± 263 
KM PRPP (µM) 182 ± 374 
kcat (s−10.57 ± 0.01 
kcat/KM ATP (s−1 mM−11.2 ± 0.07 
kcat/KM PRPP (s−1 mM−13.1 ± 0.64 
IC50 (His) (µM) 30.1 ± 2.6 
IC50 (AMP) (mM) Not detectable5 
IC50 (His + 2 mM AMP) (µM) 6.1 ± 0.56 
Parameter MedtrATP-PRT1 
KM ATP (µM) 470 ± 263 
KM PRPP (µM) 182 ± 374 
kcat (s−10.57 ± 0.01 
kcat/KM ATP (s−1 mM−11.2 ± 0.07 
kcat/KM PRPP (s−1 mM−13.1 ± 0.64 
IC50 (His) (µM) 30.1 ± 2.6 
IC50 (AMP) (mM) Not detectable5 
IC50 (His + 2 mM AMP) (µM) 6.1 ± 0.56 
1

Apparent KM value determined at 1 mM PRPP.

2

Apparent KM value determined at 2 mM ATP.

3

Determined using 0–8 mM AMP.

The crystal structure of MedtrATP-PRT1 was solved de novo using the SAD method. In both structures presented herein that are MedtrATP-PRT1 in the relaxed (R-MedtrATP-PRT1) and tense (T-MedtrATP-PRT1) states, there is one homohexamer in the asymmetric unit. Homohexamer is the oligomeric state observed on size exclusion chromatography (not shown) which is also in accord with the result of PISA analysis [61], based on the evaluation of inter-subunit contacts within the crystal lattice. Per hexamer of a surface area measuring ∼88 000 Å2, ∼23 000 Å2 of the protein surface is buried, when compared with the total surface of six isolated subunits (111 000 Å2). The MedtrATP-PRT1 homohexamer is a dimer of trimers (32 symmetry; chains A, B, C, and D, E, F; Figure 2), as the interface within the trimer is ∼2600 Å2 per subunit, and within the dimer, ∼1700 Å2. Notably, individual subunits are relatively flexible, which is reflected by the rmsd values between the chains of R-MedtrATP-PRT1 (refined without NCS restraints) that vary between 0.3 and 0.7 Å.

Structure of the MedtrATP-PRT1 homohexamer in the R-state.

Figure 2.
Structure of the MedtrATP-PRT1 homohexamer in the R-state.

Surface representation is shown in (A), whereas (B) illustrates secondary structure elements. Subunits of the hexamer are marked by circled (A–F).

Figure 2.
Structure of the MedtrATP-PRT1 homohexamer in the R-state.

Surface representation is shown in (A), whereas (B) illustrates secondary structure elements. Subunits of the hexamer are marked by circled (A–F).

The MedtrATP-PRT1 subunit comprises twelve α-helices and seventeen β-strands that are arranged in three domains (I, II, and III) (Figure 3). Domains I (residues 36–133, 237–272, and 360–373) and II (134–236) together shape a bilobate catalytic core, named as such per analogy to the structures of bacterial ATP-PRTs [60]. The regulatory domain III (residues 273–359) has a ferredoxin-like fold (βαββαβ) and in a three-way manner forms an apex of the quaternary assembly. The antiparallel β-sheet of the regulatory domain of subunit A is extended by the β17* (* denotes a structural element from the next subunit, in a clockwise order) strand of subunit B. Because strands β17 are swapped between the subunits in a three-wise manner, a solitary domain III—hence a monomeric or a dimeric MedtrATP-PRT1—is very unlikely to exist in solution. This is consistent with the current conclusions regarding bacterial ATP-PRTs [29,30], which contradicted former assumptions that the enzymes were dimers in active states [28]. The structure of MedtrATP-PRT1 revealed that the general protein architecture resembles the bacterial long-type ATP-PRTs, despite sequence identity of only ∼30% between plant and bacterial orthologs. Furthermore, since plant ATP-PRTs are conserved (e.g. 82.6% identity with AtATP-PRT1), orthologs from other plants species will probably exhibit the same architecture.

Structure of the MedtrATP-PRT1 subunit in the R-state.

Figure 3.
Structure of the MedtrATP-PRT1 subunit in the R-state.

Domains within the subunit are color-coded: domain I, pale yellow; domain II, turquoise; domain III (regulatory), violet. The β17* strand (orange), which completes the β-sheet of the regulatory domain, is from another subunit.

Figure 3.
Structure of the MedtrATP-PRT1 subunit in the R-state.

Domains within the subunit are color-coded: domain I, pale yellow; domain II, turquoise; domain III (regulatory), violet. The β17* strand (orange), which completes the β-sheet of the regulatory domain, is from another subunit.

MedtrATP-PRT1 contains features that discriminate it from bacterial orthologs and a conserved active site

Having the structure of MedtrATP-PRT1, PDBeFOLD [62] served to search for the most similar ATP-PRTs in the entire Protein Data Bank (PDB) [63]. Excluding structures of bare core domains or those of short-type ATP-PRTs, by rmsd (and by sequence identity) the closest neighbor of R-MedtrATP-PRT1 (chain A) with a known structure is CjATP-PRT (PDB ID: 4yb7, chain G [29], rmsd 1.4 Å between 189 Cα pairs, 30% sequence identity, 52% similarity). The majority of the secondary structure elements align well, although the catalytic cores and regulatory domains have to be analyzed separately (Figure 4A,B, respectively). The most outstanding exception is the presence of an additional loop and a helix (α12) in the C-terminal region of MedtrATP-PRT1 (Figure 4A). Overall, C-terminus of MedtrATP-PRT1 is longer by sixteen residues (358–373) than its counterpart in CjATP-PRT or in other bacterial sources with known 3-D structures. Moreover, based on the sequence similarity, it is very likely that the At orthologs (Figure 1B), and possibly other ATP-PRTs of plant origin, contain such an additional helix as well. The extra α12 helix interacts with the helix α9 and the strand β5 of the same subunit, which makes it a part of domain I. As a result, the hinge region (residues 271–273 and 353–354) is fastened by the helix α12, forming an anchor which secures the regulatory domain to the catalytic core. This molecular grip is also a very plausible explanation to the different conformation of the regulatory domains of R-MedtrATP-PRT1 and R-CjATP-PRT (G344 Cα in R-MedtrATP-PRT1 is 8 Å away from A285 Cα in R-CjATP-PRT; Figure 4A) even though their catalytic cores align well (Figure 4A). Notably, in bacterial R-state ATP-PRTs, which all lack the C-terminal helix (PDB IDs: 1h3d [27], 1nh7 [28], and 4yb7 [29]), positions of the regulatory domains differ by no more than ∼2.2 Å (not shown).

Comparison of MedtrATP-PRT1 with other ATP-PRTs.

Figure 4.
Comparison of MedtrATP-PRT1 with other ATP-PRTs.

Superposition of MedtrATP-PRT1 (gray) and CjATP-PRT (orange); the catalytic cores and regulatory domains are superposed in (A,B), respectively; fragments that superpose well are semitransparent; α12 helix is dark gray. Analysis of residue conservation is shown on the surface of MedtrATP-PRT1 subunit in (C), and on ribbons in (D). A chloride anion (from R-MedtrATP-PRT1 structure) and PR-ATP (superposed from CjATP-PRT catalytic core complex with PR-ATP; PDB ID: 5ubg [60]) are shown in (D); conserved residues of the regulatory domain are indicated.

Figure 4.
Comparison of MedtrATP-PRT1 with other ATP-PRTs.

Superposition of MedtrATP-PRT1 (gray) and CjATP-PRT (orange); the catalytic cores and regulatory domains are superposed in (A,B), respectively; fragments that superpose well are semitransparent; α12 helix is dark gray. Analysis of residue conservation is shown on the surface of MedtrATP-PRT1 subunit in (C), and on ribbons in (D). A chloride anion (from R-MedtrATP-PRT1 structure) and PR-ATP (superposed from CjATP-PRT catalytic core complex with PR-ATP; PDB ID: 5ubg [60]) are shown in (D); conserved residues of the regulatory domain are indicated.

Another noticeable difference concerns the helix α3 of MedtrATP-PRT1 which is by five residues shorter than its equivalent of CjATP-PRT (Figure 4A). This fragment in R-CjATP-PRT points towards the solvent/cytoplasm, but in T-CjATP-PRT the helices belonging to two subunits are shifted towards each other [29]. Interestingly, in EcATP-PRT the fragment is shorter than in CjATP-PRT, thus resembles the plant protein (not shown). Next, MedtrATP-PRT1 contains the helix α8, whereas the corresponding fragment of CjATP-PRT is a short loop. When the regulatory domains are superposed, the helix α10 has a different inclination with respect to the rest of the domain (Figure 4B), whereas respective elements of the catalytic core are positioned up to 16 Å away from each other. Finally, MedtrATP-PRT1 lacks disulfide bridges, and in fact, there are no cysteine residues positioned to form S-S bonds. This is in contrast to e.g. EcATP-PRT, wherein a disulfide bond between Cys73 and Cys175 was present in the R-state but missing in the T-state, and was proposed to have a potential role in activity regulation [28].

Analysis of the residue conservation with ConSurf [53] revealed invariable patches on the MedtrATP-PRT1 surface that is elsewhere rather variable (Figure 4C). One of those patches maps to the crevice between the domains I and II, a region which is equivalent to the active site of bacterial ATP-PRTs [27,28]. Superposition of R-MedtrATP-PRT1 with a recent structure of the CjATP-PRT catalytic core (without the regulatory domain, PDB ID: 5ubg, chain A [60]; rmsd 1.4 Å between 175 Cα pairs) in complex with the product, PR-ATP was performed to propose functions of the conserved active-site regions. Based on this comparative analysis, domain I of R-MedtrATP-PRT1, and more precisely N-termini of the helices α1, α2, and α3 overlap with the CjATP-PRT site binding the ATP moiety of PR-ATP (Figure 4D). The phosphoribosyl moiety binding site corresponds to the loop preceding and the N-terminus of the helix α6, and the loop between the strand β9 and the helix α7, which agrees with the canonical position of the PRPP loop in PRT enzymes in general [64]. In the R-MedtrATP-PRT1 structure, a chloride anion is bound at the position overlapping with the 5′-phosphate of PRPP (Figure 4D). The presence of chloride results from the high NaCl concentration in the crystallization solution, but undoubtedly corroborates this fragment as a placeholder for anions. Moreover, the proposed binding sites for both ATP and phosphoribosyl moieties are in an agreement with the binary complexes of CjATP-PRT with either of the two reaction substrates (PDB ID: 5ubh, and 5ubi for ATP and PRPP complexes, respectively [60]).

The allosteric histidine-binding site has a slightly different environment than in bacterial ATP-PRTs

Inspection of the electron density maps of the complex obtained in the presence of histidine revealed six histidine molecules, each allosterically bound at the inter-subunit interface of the regulatory domains (Figure 5A,B). Histidine is secured via an extensive network of hydrogen bonds contributed by the regulatory domains of two subunits (Figure 5B). The histidine carboxyl group is hydrogen bonded to backbone amides of Leu301, Gln302, and Cys327, and to Nδ of Asn279** (** indicates an element of a preceding subunit—in a clockwise manner). Asn279**, by its Oδ atom also interacts with the histidine amine that is in turn H-bonded to the carbonyl of Gly303 and Oγ of Thr305. Finally, the imidazole moiety of histidine, by its Nε atom, binds to the carbonyl of Gly347**. Thus, in each case regulatory domain of the preceding subunit (or two subunits away, clockwise) contributes to histidine binding but, as already stated, the β-sheet of each regulatory domain is extended by the β17 strand of the succeeding subunit.

Allosteric binding of histidine.

Figure 5.
Allosteric binding of histidine.

Three sites of histidine (spheres) binding, on one face of the MedtrATP-PRT1 hexamer in T-state are presented in (A); part of the protein surface is clipped and ribbons are semitransparent. (B) The detailed histidine-binding mode. Fo− Fc omit map (green mesh) for histidine is contoured at 3σ level. Double asterisks (**) denote residues from a preceding subunit in a clockwise order. Histidine binding by CjATP-PRT is presented in (C). (D) Residue conservation of the regulatory domain of MedtrATP-PRT1 color-coded according to the scale on the right.

Figure 5.
Allosteric binding of histidine.

Three sites of histidine (spheres) binding, on one face of the MedtrATP-PRT1 hexamer in T-state are presented in (A); part of the protein surface is clipped and ribbons are semitransparent. (B) The detailed histidine-binding mode. Fo− Fc omit map (green mesh) for histidine is contoured at 3σ level. Double asterisks (**) denote residues from a preceding subunit in a clockwise order. Histidine binding by CjATP-PRT is presented in (C). (D) Residue conservation of the regulatory domain of MedtrATP-PRT1 color-coded according to the scale on the right.

Although the location of the histidine binding site in T-MedtrATP-PRT1 is similar to that of the prokaryotic enzymes with reported structures, the environment inside the binding cavity is somewhat different. First, Asn279** of MedtrATP-PRT1 is not conserved in prokaryotic examples (Figure 1B). For instance, it is substituted by a histidine residue in CjATP-PRT (His232**, Figure 5C) or by an aspartate (Asp218) in MtATP-PRT, which indeed have been shown to interact with the histidine amino group [28,29]. The second difference is a lack of water-mediated interaction in T-MedtrATP-PRT1, although a weak peak of positive electron density might suggest the presence of a water molecule at a partial occupancy (only 20 water molecules in total were modeled due to low resolution). Nonetheless, among the three residues that hold the water in place in CjATP-PRT (Tyr228, Thr252, and His266**; PDB ID: 4yb6 [29]), only the threonine residue is conserved in MedtrATP-PRT1 (Thr305, Figure 5B,C). On the other hand, all residues which interact with histidine in MedtrATP-PRT1 are conserved in both At isoforms (Figure 1B). Considering a larger subset of ATP-PRT sequences, the overall conservation of residues in the regulatory domains is rather low, and only Asn279, Pro304, Thr305, Ala323, Leu340, Gly344, Gly345, and Val348 are relatively invariable (Figure 5D). These residues are arranged across the regulatory domain (Figure 4D) and are located close to the histidine binding sites at the interfaces with either of the two neighboring subunits. However, it must also be noted that most of the histidine-binding interactions are contributed by the protein backbone atoms, including the conserved 300GXXXPT305 motif, which ensures similar positioning of histidine despite the high sequence variability.

Histidine binding causes a transition from R-state to T-state that involves a trans- to cis-peptide switch

Similar to the behavior of bacterial ATP-PRTs, histidine binding by MedtrATP-PRT1 causes substantial, rigid-body concomitant movements of domains belonging to the six subunits. As a result, the protein undergoes a transition from R-state to T-state. Change of the conformational state is attained by the regulatory domains, which force a twist of the catalytic cores. As a result, external dimensions of the protein change from ∼128 × 117 × 106 Å to 138 × 109 × 100 Å (Figure 6), hence T-MedtrATP-PRT1 is more elongated in contrast to the bulkier R-MedtrATP-PRT1. Both states of MedtrATP-PRT1 are significantly larger than their bacterial counterparts; for instance, R-MtATP-PRT measures 120 × 105 × 98 Å, whereas T-MtATP-PRT is 125 × 98 × 94 Å (based on PDB IDs 1nh7 and 1nh8, respectively [28]). The transition from R-MedtrATP-PRT1 to T-MedtrATP-PRT1 does not appear to impact the overall distribution of electrostatic potential on either the outer (negatively charged) or inner (within the void, positive) surface of the hexamer (Figure 6). However, the void in the core of T-MedtrATP-PRT1 is smaller and has a different shape than that of R-MedtrATP-PRT1 (Figure 6).

MedtrATP-PRT1 hexamer in R-state (A) and T-state (B).

Figure 6.
MedtrATP-PRT1 hexamer in R-state (A) and T-state (B).

Electrostatic potential is mapped on the surface. External dimensions of the hexamer are given in the left panels; clearance change of the opening that leads to the hexamer void is marked by semitransparent ovals. Panels to the right show clipped surfaces to illustrate changes in the lumen of the hexamer.

Figure 6.
MedtrATP-PRT1 hexamer in R-state (A) and T-state (B).

Electrostatic potential is mapped on the surface. External dimensions of the hexamer are given in the left panels; clearance change of the opening that leads to the hexamer void is marked by semitransparent ovals. Panels to the right show clipped surfaces to illustrate changes in the lumen of the hexamer.

The regulatory domains themselves do not change drastically, as rearrangements are observed only near the histidine binding sites. However, when the regulatory domains of R-MedtrATP-PRT1 and T-MedtrATP-PRT1 are superposed, the catalytic cores lie 14–15 Å away from each other (Figure 7A). Interestingly, in MtATP-PRT—and other bacterial ATP-PRTs—the R-to-T-state transition was accompanied by a movement of up to 30 Å (PDB IDs 1nh7 and 1nh8, respectively [28]). When catalytic cores of the R- and T-states of both MedtrATP-PRT1 and MtATP-PRT are superposed, it becomes clear that all four conformations of regulatory domains (R-MedtrATP-PRT1, T-MedtrATP-PRT1, R-MtATP-PRT, and T-MtATP-PRT) are different. In this view, the regulatory domain in T-MedtrATP-PRT1 appears to be conformationally extreme when relative positioning with respect to the catalytic cores is considered (Figure 7B). Such an extreme twist carries serious structural consequences within the regions between the catalytic cores and regulatory domains. In particular, a cis-peptide is present between Met271 and Gly272 in T-MedtrATP-PRT1 (Figure 7C,D). A case of trans- to cis-peptide switch at the catalytic core-regulatory domain interface has not been observed in bacterial ATP-PRTs. It is possible that the switch occurs because the helix α12, which is unique to plant ATP-PRTs, provides an additional grip that impacts the conformation of the domains. Consistently, upon the transition, the helix α12 maintains the position relative to the catalytic core, and not to the regulatory domain (not shown). It is thus likely that the peptide switch is salient to plant ATP-PRTs.

A detailed view on the transition from R- to T-state.

Figure 7.
A detailed view on the transition from R- to T-state.

(A) Positioning of the catalytic cores of the MedtrATP-PRT1 subunits in the two states when the regulatory domains are superposed; helices α12 are shown as ribbons, rest of the model as licorice. (B) Comparison of the positioning of the regulatory domains with respect to the catalytic cores of the two states of MedtrATP-PRT1 and MtATP-PRT; only the helix α11 (MedtrATP-PRT1 topology) is shown as ribbons. Note the extreme conformation of T-MedtrATP-PRT1. (C,D) Conformational change in the hinge region made up by two peptide strands. Blue mesh shows 2Fo − Fc electron density maps for residues 266–274, and 351–355 at 1.5σ level. Protein chains are color-coded according to the legend.

Figure 7.
A detailed view on the transition from R- to T-state.

(A) Positioning of the catalytic cores of the MedtrATP-PRT1 subunits in the two states when the regulatory domains are superposed; helices α12 are shown as ribbons, rest of the model as licorice. (B) Comparison of the positioning of the regulatory domains with respect to the catalytic cores of the two states of MedtrATP-PRT1 and MtATP-PRT; only the helix α11 (MedtrATP-PRT1 topology) is shown as ribbons. Note the extreme conformation of T-MedtrATP-PRT1. (C,D) Conformational change in the hinge region made up by two peptide strands. Blue mesh shows 2Fo − Fc electron density maps for residues 266–274, and 351–355 at 1.5σ level. Protein chains are color-coded according to the legend.

The R- to T-state transition changes clearance of the largest channel leading to the hexamer void from ∼19 × 9 Å to 16 × 6 Å (Figure 6). Notably, in T-MedtrATP-PRT1, another small opening, of 15 × 3 Å becomes available (not shown). Larger of the two ATP-PRT substrates, ATP, measures ∼18 Å in its most elongated conformation, which indicates that decreasing the opening size may be related with histidine inhibition. Furthermore, the product, PR-ATP is even larger and therefore its release from the void appears to be difficult in the T-state. It remains to be determined if closing the channels is directly the basis of inhibition in the case of MedtrATP-PRT1. Using pre-steady-state kinetics on a bacterial ortholog, MtATP-PRT, it was shown that chemistry is not the rate-limiting step because burst kinetics were observed [30]. The authors also excluded slower product release (from the catalytic site, but not from the void) as the inhibition mechanism on the basis of the steady-state rate which did not decrease. Hence, histidine inhibition was linked to trapping the enzyme in an inactive complex [30]. With this in mind, however, it would still be possible that the inhibited enzyme could catalyze the phosphoribosyl transfer, but accumulation of the product (PR-ATP) in the void might cause steric hindrances and ultimately inhibit the enzyme. However, more recent studies (also on MtATP-PRT) by Pisco et al. [65] have shown that the enzyme can be activated by 3-(2-thienyl)-L-alanine (TIH), but this complex is in the T-state. This suggested that the activity and/or inhibition is controlled not by a static distribution of R- versus T-state enzymes but rather by dynamic changes from the R-to-T-to-R-state during the course of the reaction. In other words, only the R-state can bind the substrates and release the product, but since the TIH-activated enzyme adopts the T-state, the authors concluded that inhibition is attributed to a sampling rate between the R- and T-states [65]. Moreover, even in the absence of histidine, ∼12% of MtATP-PRT was shown to be in T-state, but R-state was not observed after adding two equivalents of histidine per site [31]. This hypothesis is also consistent with the current model of allosteric enzyme regulation in general, whereby a broader spectrum of enzyme conformations are tunable, and allosteric regulators impact the energy landscape of the enzymes [66]. In this model, the relative distribution of multiple states is the basis of regulation. Judging by the kinetic parameters that are similar, plant [32] and the extensively studied bacterial ATP-PRTs may be alike in the context of allosteric regulation. Hopefully, with the structural knowledge, further studies will reveal the exact nature of the allosteric regulation of plant ATP-PRTs.

AMP binding is consistent with competitive inhibition

AMP is a known competitive inhibitor of bacterial ATP-PRTs towards both substrates, ATP and PRPP, and acts synergistically, increasing the inhibition by histidine alone [33]. In contrast, MedtrATP-PRT1 is not inhibited by AMP, although AMP still increases sensitivity to histidine (Figure 1D). In the T-MedtrATP-PRT1 complex with histidine and AMP, six molecules of AMP were found, each occupying an equivalent position, consistent with the active site of ATP-PRTs. In T-MedtrATP-PRT1, the AMP phosphate group is held in place via multiple hydrogen bonds donated by the 215SSGTT219 fragment (Figure 8A), which has been previously described as the PRPP binding loop [27,28]. Notably, the location of the AMP phosphate overlaps with Cl- in this work R-MedtrATP-PRT1 structure (Figure 4D). The adenine moiety is stacked with Leu213 and interacts with Sγ of Cys134 by a sulfur-aromatic type of contact [67]. The closest residue able to bind the AMP ribose (Asp212) is ∼3.3 Å away from the O2′ atom (Figure 8A) thus this interaction appears to be rather weak. On the other hand, Asp212 is shifted by ∼1.8 Å towards the AMP ribose when compared with R-MedtrATP-PRT1, indicating that the H-bonding does occur.

AMP binding within the active site.

Figure 8.
AMP binding within the active site.

(A) The Fo − Fc omit map (green mesh) around AMP is contoured at 3σ level. AMP-binding modes observed in MedtrATP-PRT1 and MtATP-PRT are shown in (B). (C) Comparison of MedtrATP-PRT1/AMP complex (AMP is gray) with position of PR-ATP (black C-atoms) in complex with the CjATP-PRT catalytic core (PDB ID: 5ubg [60]).

Figure 8.
AMP binding within the active site.

(A) The Fo − Fc omit map (green mesh) around AMP is contoured at 3σ level. AMP-binding modes observed in MedtrATP-PRT1 and MtATP-PRT are shown in (B). (C) Comparison of MedtrATP-PRT1/AMP complex (AMP is gray) with position of PR-ATP (black C-atoms) in complex with the CjATP-PRT catalytic core (PDB ID: 5ubg [60]).

Overall, AMP binding by T-MedtrATP-PRT1 resembles that observed in EcATP-PRT (PDB ID: 1h3d [27]) and CjATP-PRT (4yb6 [29]) (not shown). In contrast, the AMP conformation is different in the complex with MtATP-PRT (1nh8 [28]) wherein only the phosphate position overlaps, but adenosine is oriented differently (Figure 8B). It is possible that AMP binding modes, inconsistent among different structures, reflect the dynamics of ATP-PRT inhibition by AMP. What is consistent, however, is the positioning of AMP different from either ATP poised for the reaction or the ATP moiety of PR-ATP (Figure 8C). Unfortunately, it was not possible to obtain complexes of MedtrATP-PRT1 with ATP, PRPP or PR-ATP. Nonetheless, superposition of T-MedtrATP-PRT1/AMP complex with the structure of the CjATP-PRT catalytic core with PR-ATP (5ubg) [60] suggests that AMP most likely occupies binding sites of both substrates and product also in the case of MedtrATP-PRT1 (Figure 8C). The observation that MedtrATP-PRT1 is insensitive to AMP inhibition, but AMP binding within the catalytic site enhances the histidine perception is consistent with the hypothesis that inhibition of ATP-PRTs is not static, and instead is based on dynamic R-to-T-to-R-state switching during the catalytic event [65]. Such a synergistic action of the two inhibitors apparently impacts the spectrum of R- and T-states of MedtrATP-PRT1.

Conclusions and future outlook

MedtrATP-PRT1 is the first example of ATP-PRT from a plant source (or a eukaryotic organism) to have its structure determined. The structure was solved using the SAD method on the SeMet-labeled protein. The crystal structures show MedtrATP-PRT1 hexamer in the R-state and, in the histidine-bound, T-state. The conformational changes associated with the R-to-T transition involve a flip of the peptide backbone between Met271 and Gly272 that adopts cis-conformation in the T-state. Very likely, such a profound change, not observed in bacterial ATP-PRTs, is related to the presence of the α12 helix, which prokaryotic examples lack. In fact, the conformations of regulatory domains with respect to the catalytic cores, which were preserved in respective R- or T-structures of the bacterial examples, are meaningfully different in the corresponding states of MedtrATP-PRT1. Furthermore, AMP is shown to augment the histidine sensitivity and bind at the active site which allows cells to control the metabolically-expensive histidine production in response to changes in their metabolic status, reflected by ATP/AMP ratio.

Studying the plant pathway of histidine biosynthesis has a potential three-fold application, in addition to expanding the basic knowledge of plant metabolism. First, because the histidine-biosynthetic pathway is absent in mammals, it is a good target for the herbicide design. Glyphosate resistance is becoming an alarming threat to the global food production, and the rising CO2 levels in the atmosphere may escalate the scale of the problem [68]. Structural information is necessary for a rationalized drug design, thus structures of enzymes that are auspicious subjects are needed for development of highly specific, ATP-PRT-targeted inhibitors against bacteria, fungi, and plants.

Second, by being able to enhance the histidine production in plants, for example by alleviating the feedback inhibition by histidine, we might be able to produce crops with a higher histidine content. In microbes, elimination of the feedback inhibition was recently studied in Corynebacterium glutamicum, with the aim to use this microorganism for histidine production [69]. Since histidine is an essential amino acid to humans, the interest is both humanitarian and economical [70]. Especially in developing countries, where plants constitute a majority of the food, enriching the content of essential amino acids in crops is desirable. In developed regions of the world, where significantly more meat products are consumed, a shortage of essential amino acids in the human diet is hardly an issue. Nonetheless, the forage production industry would undoubtedly benefit from such improved crops.

Third, because free histidine pool resulting from a hyperactivated ATP-PRT has been shown to increase e.g. nickel tolerance in A. thaliana [71], plants with a high histidine content could be used to remove heavy metal contamination from soil. The structures of MedtrATP-PRT1 presented in this work are certainly a necessary step towards the three aforementioned applications related to histidine biosynthesis in plants.

Database Depositions

PDB codes: MedtrATP-PRT1 in the relaxed state: 6czl; MedtrATP-PRT1 in the tense state (complex with histidine and AMP): 6czm.

Abbreviations

     
  • AMP

    adenosine 5′-monophosphate

  •  
  • At

    Arabidopsis thaliana

  •  
  • ATP

    adenosine 5′-triphosphate

  •  
  • ATP-PRT

    ATP-phosphoribosyltransferase

  •  
  • NCS

    non-crystallographic symmetry

  •  
  • PDB

    Protein Data Bank

  •  
  • PR-ATP

    phosphoribosyl-ATP

  •  
  • PRPP

    5-phosphoribosyl-α1-pyrophosphate

  •  
  • SAD

    single anomalous dispersion

  •  
  • SeMet

    selenomethionine

  •  
  • TCEP

    tris(2-carboxyethyl)phosphine

Acknowledgments

The author is extremely grateful to Zbigniew Dauter (NCI) for his guidance and supervision, and to Jessica Johnson (Argonne) for help with the Biotek spectrophotometer setup. This project was supported by the Intramural Research Program of the NCI Center for Cancer Research. Diffraction data were collected at the SER-CAT beamline 22-ID at the Advanced Photon Source, Argonne National Laboratory, supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Contract W-31-109-Eng-38, and beamline 19-ID at the Structural Biology Center at the Advanced Photon Source, operated by UChicago Argonne, LLC, for the U.S. Department of Energy, Office of Biological and Environmental Research under contract DE-AC02-06CH11357.

Competing Interests

The Author declares that there are no competing interests associated with this manuscript.

References

References
1
Lunardi
,
J.
,
Nunes
,
J.E.
,
Bizarro
,
C.V.
,
Basso
,
L.A.
,
Santos
,
D.S.
and
Machado
,
P.
(
2013
)
Targeting the histidine pathway in Mycobacterium tuberculosis
.
Curr. Top. Med. Chem.
13
,
2866
2884
2
Jastrzębowska
,
K.
and
Gabriel
,
I.
(
2015
)
Inhibitors of amino acids biosynthesis as antifungal agents
.
Amino Acids
47
,
227
249
3
Mousdale
,
D. M.
and
Coggins
,
J. R.
(
1991
) Amino acid synthesis. In
Target Sites for Herbicide Action
(
Kirkwood
,
R. C.
ed.). pp.
29
56
,
Plenum Press
,
New York
4
Ingle
,
R.A.
(
2011
)
Histidine biosynthesis
.
Arabidopsis Book
9
,
e0141
5
Stepansky
,
A.
and
Leustek
,
T.
(
2006
)
Histidine biosynthesis in plants
.
Amino Acids
30
,
127
142
6
Wiater
,
A.
,
Krajewska-Grynkiewicz
,
K.
and
Klopotowski
,
T.
(
1971
)
Histidine biosynthesis and its regulation in higher plants
.
Acta Biochim. Pol.
18
,
299
307
PMID:
[PubMed]
7
Fujimori
,
K.
and
Ohta
,
D.
(
1998
)
Isolation and characterization of a histidine biosynthetic gene in Arabidopsis encoding a polypeptide with two separate domains for phosphoribosyl-ATP pyrophosphohydrolase and phosphoribosyl-AMP cyclohydrolase
.
Plant Physiol.
118
,
275
283
8
Fujimori
,
K.
,
Tada
,
S.
,
Kanai
,
S.
and
Ohta
,
D.
(
1998
)
Molecular cloning and characterization of the gene encoding N′-[(5′-phosphoribosyl)-formimino]-5-aminoimidazole-4-carboxamide ribonucleotide (BBM II) isomerase from Arabidopsis thaliana
.
Mol. Gen. Genet.
259
,
216
223
9
Fujimori
,
K.
and
Ohta
,
D.
(
1998
)
An Arabidopsis cDNA encoding a bifunctional glutamine amidotransferase/cyclase suppresses the histidine auxotrophy of a Saccharomyces cerevisiae his7 mutant
.
FEBS Lett.
428
,
229
234
10
El Malki
,
F.
,
Frankard
,
V.
and
Jacobs
,
M.
(
1998
)
Molecular cloning and expression of a cDNA sequence encoding histidinol phosphate aminotransferase from Nicotiana tabacum
.
Plant Mol. Biol.
37
,
1013
1022
11
Petersen
,
L.N.
,
Marineo
,
S.
,
Mandala
,
S.
,
Davids
,
F.
,
Sewell
,
B.T.
and
Ingle
,
R.A.
(
2010
)
The missing link in plant histidine biosynthesis: Arabidopsis myoinositol monophosphatase-like2 encodes a functional histidinol-phosphate phosphatase
.
Plant Physiol.
152
,
1186
1196
12
Wong
,
Y.S.
and
Mazelis
,
M.
(
1981
)
Detection and properties of L-histidinol dehydrogenase in wheat germ
.
Phytochemistry
20
,
1831
1834
13
Nagai
,
A.
,
Ward
,
E.
,
Beck
,
J.
,
Tada
,
S.
,
Chang
,
J.Y.
,
Scheidegger
,
A.
et al. 
(
1991
)
Structural and functional conservation of histidinol dehydrogenase between plants and microbes
.
Proc. Natl Acad. Sci. U.S.A.
88
,
4133
4137
14
Nagai
,
A.
and
Scheidegger
,
A.
(
1991
)
Purification and characterization of histidinol dehydrogenase from cabbage
.
Arch. Biochem. Biophys.
284
,
127
132
15
Bisson
,
C.
,
Britton
,
K.L.
,
Sedelnikova
,
S.E.
,
Rodgers
,
H.F.
,
Eadsforth
,
T.C.
,
Viner
,
R.C.
et al. 
(
2015
)
Crystal structures reveal that the reaction mechanism of imidazoleglycerol-phosphate dehydratase is controlled by switching Mn(II) coordination
.
Structure
23
,
1236
1245
16
Glynn
,
S.E.
,
Baker
,
P.J.
,
Sedelnikova
,
S.E.
,
Davies
,
C.L.
,
Eadsforth
,
T.C.
,
Levy
,
C.W.
et al. 
(
2005
)
Structure and mechanism of imidazoleglycerol-phosphate dehydratase
.
Structure
13
,
1809
1817
17
Ruszkowski
,
M.
and
Dauter
,
Z.
(
2016
)
Structural studies of Medicago truncatula histidinol phosphate phosphatase from inositol monophosphatase superfamily reveal details of penultimate step of histidine biosynthesis in plants
.
J. Biol. Chem.
291
,
9960
9973
18
Ruszkowski
,
M.
and
Dauter
,
Z.
(
2017
)
Structures of Medicago truncatula L-histidinol dehydrogenase show rearrangements required for NAD+ binding and the cofactor positioned to accept a hydride
.
Sci. Rep.
7
,
10476
19
Voll
,
M.J.
,
Appella
,
E.
and
Martin
,
R.G.
(
1967
)
Purification and composition studies of phosphoribosyladenosine triphosphate:pyrophosphate phosphoribosyltransferase, the first enzyme of histidine biosynthesis
.
J. Biol. Chem.
242
,
1760
1767
PMID:
[PubMed]
20
Ames
,
B.N.
,
Martin
,
R.G.
and
Garry
,
B.J.
(
1961
)
The first step of histidine biosynthesis
.
J. Biol. Chem.
236
,
2019
2026
PMID:
[PubMed]
21
Morton
,
D.P.
and
Parsons
,
S.M.
(
1976
)
Biosynthetic direction substrate kinetics and product inhibition studies on the first enzyme of histidine biosynthesis, adenosine triphosphate phosphoribosyltransferase
.
Arch. Biochem. Biophys.
175
,
677
686
22
Brashear
,
W.T.
and
Parsons
,
S.M.
(
1975
)
Evidence against a covalent intermediate in the adenosine triphosphate phosphoribosyltransferase reaction of histidine biosynthesis
.
J. Biol. Chem.
250
,
6885
6890
PMID:
[PubMed]
23
Champagne
,
K.S.
,
Sissler
,
M.
,
Larrabee
,
Y.
,
Doublié
,
S.
and
Francklyn
,
C.S.
(
2005
)
Activation of the hetero-octameric ATP phosphoribosyl transferase through subunit interface rearrangement by a tRNA synthetase paralog
.
J. Biol. Chem.
280
,
34096
34104
24
Bond
,
J.P.
and
Francklyn
,
C.
(
2000
)
Proteobacterial histidine-biosynthetic pathways are paraphyletic
.
J. Mol. Evol.
50
,
339
347
25
Moggré
,
G.-J.
,
Poulin
,
M.B.
,
Tyler
,
P.C.
,
Schramm
,
V.L.
and
Parker
,
E.J.
(
2017
)
Transition state analysis of adenosine triphosphate phosphoribosyltransferase
.
ACS Chem. Biol.
12
,
2662
2670
26
Bovee
,
M.L.
,
Champagne
,
K.S.
,
Demeler
,
B.
and
Francklyn
,
C.S.
(
2002
)
The quaternary structure of the HisZ-HisG N-1-(5′-phosphoribosyl)-ATP transferase from Lactococcus lactis
.
Biochemistry
41
,
11838
11846
27
Lohkamp
,
B.
,
McDermott
,
G.
,
Campbell
,
S.A.
,
Coggins
,
J.R.
and
Lapthorn
,
A.J.
(
2004
)
The structure of Escherichia coli ATP-phosphoribosyltransferase: identification of substrate binding sites and mode of AMP inhibition
.
J. Mol. Biol.
336
,
131
144
28
Cho
,
Y.
,
Sharma
,
V.
and
Sacchettini
,
J.C.
(
2003
)
Crystal structure of ATP phosphoribosyltransferase from Mycobacterium tuberculosis
.
J. Biol. Chem.
278
,
8333
8339
29
Mittelstädt
,
G.
,
Moggré
,
G.J.
,
Panjikar
,
S.
,
Nazmi
,
A.R.
and
Parker
,
E.J.
(
2016
)
Campylobacter jejuni adenosine triphosphate phosphoribosyltransferase is an active hexamer that is allosterically controlled by the twisting of a regulatory tail
.
Protein Sci.
25
,
1492
1506
30
Pedreño
,
S.
,
Pisco
,
J.P.
,
Larrouy-Maumus
,
G.
,
Kelly
,
G.
and
de Carvalho
,
L.P.S.
(
2012
)
Mechanism of feedback allosteric inhibition of ATP phosphoribosyltransferase
.
Biochemistry
51
,
8027
8038
31
Pacholarz
,
K.J.
,
Burnley
,
R.J.
,
Jowitt
,
T.A.
,
Ordsmith
,
V.
,
Pisco
,
J.P.
,
Porrini
,
M.
et al. 
(
2017
)
Hybrid mass spectrometry approaches to determine how L-histidine feedback regulates the enzyzme MtATP-phosphoribosyltransferase
.
Structure
25
,
730
738.e4
32
Ohta
,
D.
,
Fujimori
,
K.
,
Mizutani
,
M.
,
Nakayama
,
Y.
,
Kunpaisal-Hashimoto
,
R.
,
Münzer
,
S.
et al. 
(
2000
)
Molecular cloning and characterization of ATP-phosphoribosyl transferase from Arabidopsis, a key enzyme in the histidine biosynthetic pathway
.
Plant Physiol.
122
,
907
914
33
Morton
,
D.P.
and
Parsons
,
S.M.
(
1977
)
Inhibition of ATP phosphoribosyltransferase by AMP and ADP in the absence and presence of histidine
.
Arch. Biochem. Biophys.
181
,
643
648
34
Swire
,
J.
(
2007
)
Selection on synthesis cost affects interprotein amino acid usage in all three domains of life
.
J. Mol. Evol.
64
,
558
571
35
Emanuelsson
,
O.
,
Nielsen
,
H.
,
Brunak
,
S.
and
von Heijne
,
G.
(
2000
)
Predicting subcellular localization of proteins based on their N-terminal amino acid sequence
.
J. Mol. Biol.
300
,
1005
1016
36
Nielsen
,
H.
,
Engelbrecht
,
J.
,
Brunak
,
S.
and
von Heijne
,
G.
(
1997
)
Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites
.
Protein Eng. Des. Sel.
10
,
1
6
37
Kim
,
Y.
,
Babnigg
,
G.
,
Jedrzejczak
,
R.
,
Eschenfeldt
,
W.H.
,
Li
,
H.
,
Maltseva
,
N.
et al. 
(
2011
)
High-throughput protein purification and quality assessment for crystallization
.
Methods
55
,
12
28
38
Makowska-Grzyska
,
M.
,
Kim
,
Y.
,
Maltseva
,
N.
,
Li
,
H.
,
Zhou
,
M.
,
Joachimiak
,
G.
et al. 
(
2014
)
Protein production for structural genomics using E. coli expression
.
Methods Mol. Biol.
1140
,
89
105
39
Minor
,
W.
,
Cymborowski
,
M.
,
Otwinowski
,
Z.
and
Chruszcz
,
M.
(
2006
)
HKL-3000: the integration of data reduction and structure solution--from diffraction images to an initial model in minutes
.
Acta Crystallogr., Sect. D: Biol. Crystallogr.
62
,
859
866
40
Kabsch
,
W.
(
2010
) Xds.
Acta Crystallogr., Sect. D: Biol. Crystallogr.
66
,
125
132
41
Sheldrick
,
G.M.
(
2008
)
A short history of SHELX
.
Acta Cryst. A Found. Crystallogr.
64
,
112
122
42
Terwilliger
,
T.C.
,
Grosse-Kunstleve
,
R.W.
,
Afonine
,
P.V.
,
Moriarty
,
N.W.
,
Zwart
,
P.H.
,
Hung
,
L.W.
et al. 
(
2008
)
Iterative model building, structure refinement and density modification with the PHENIX AutoBuild wizard
.
Acta Crystallogr., Sect. D: Biol. Crystallogr.
64
,
61
69
43
Kowiel
,
M.
,
Jaskolski
,
M.
and
Dauter
,
Z.
(
2014
)
ACHESYM: an algorithm and server for standardized placement of macromolecular models in the unit cell
.
Acta Crystallogr., Sect. D: Biol. Crystallogr.
70
,
3290
3298
44
Emsley
,
P.
,
Lohkamp
,
B.
,
Scott
,
W.G.
and
Cowtan
,
K.
(
2010
)
Features and development of Coot
.
Acta Crystallogr., Sect. D: Biol. Crystallogr.
66
,
486
501
45
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., Sect. D: Biol. Crystallogr.
67
,
355
367
46
Winn
,
M.D.
,
Murshudov
,
G.N.
and
Papiz
,
M.Z.
(
2003
)
Macromolecular TLS refinement in REFMAC at moderate resolutions
.
Methods Enzymol.
374
,
300
321
47
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
,
658
674
48
Headd
,
J.J.
,
Echols
,
N.
,
Afonine
,
P.V.
,
Moriarty
,
N.W.
,
Gildea
,
R.J.
and
Adams
,
P.D.
(
2014
)
Flexible torsion-angle noncrystallographic symmetry restraints for improved macromolecular structure refinement
.
Acta Crystallogr., Sect. D: Biol. Crystallogr.
70
,
1346
1356
49
Afonine
,
P.V.
,
Grosse-Kunstleve
,
R.W.
,
Echols
,
N.
,
Headd
,
J.J.
,
Moriarty
,
N.W.
,
Mustyakimov
,
M.
et al. 
(
2012
)
Towards automated crystallographic structure refinement with phenix.refine
.
Acta Crystallogr., Sect. D: Biol. Crystallogr.
68
,
352
367
50
Heldt
,
W.H.
,
Werdan
,
K.
,
Milovancev
,
M.
and
Geller
,
G.
(
1973
)
Alkalization of the chloroplast stroma caused by light-dependent proton flux into the thylakoid space
.
Biochim. Biophys. Acta, Bioenerg.
314
,
224
241
51
Pettersen
,
E.F.
,
Goddard
,
T.D.
,
Huang
,
C.C.
,
Couch
,
G.S.
,
Greenblatt
,
D.M.
,
Meng
,
E.C.
et al. 
(
2004
)
UCSF Chimera-a visualization system for exploratory research and analysis
.
J. Comput. Chem.
25
,
1605
1612
52
Laskowski
,
R.A.
,
Jabłońska
,
J.
,
Pravda
,
L.
,
Vařeková
,
R.S.
and
Thornton
,
J.M.
(
2018
)
PDBsum: structural summaries of PDB entries
.
Protein Sci.
27
,
129
134
53
Ashkenazy
,
H.
,
Abadi
,
S.
,
Martz
,
E.
,
Chay
,
O.
,
Mayrose
,
I.
,
Pupko
,
T.
et al. 
(
2016
)
Consurf 2016: an improved methodology to estimate and visualize evolutionary conservation in macromolecules
.
Nucleic Acids Res.
44
,
W344
W350
54
The UniProt, C.
(
2017
)
Uniprot: the universal protein knowledgebase
.
Nucleic Acids Res.
45
,
D158
D169
55
Finn
,
R.D.
,
Clements
,
J.
and
Eddy
,
S.R.
(
2011
)
HMMER web server: interactive sequence similarity searching
.
Nucleic Acids Res.
39
,
W29
W37
56
Dolinsky
,
T.J.
,
Nielsen
,
J.E.
,
McCammon
,
J.A.
and
Baker
,
N.A.
(
2004
)
PDB2PQR: an automated pipeline for the setup of Poisson-Boltzmann electrostatics calculations
.
Nucleic Acids Res.
32
,
W665
W667
57
Baker
,
N.A.
,
Sept
,
D.
,
Joseph
,
S.
,
Holst
,
M.J.
and
McCammon
,
J.A.
(
2001
)
Electrostatics of nanosystems: application to microtubules and the ribosome
.
Proc. Natl Acad. Sci. U.S.A.
98
,
10037
10041
58
Altschul
,
S.F.
,
Gish
,
W.
,
Miller
,
W.
,
Myers
,
E.W.
and
Lipman
,
D.J.
(
1990
)
Basic local alignment search tool
.
J. Mol. Biol.
215
,
403
410
59
Altschul
,
S.F.
,
Madden
,
T.L.
,
Schaffer
,
A.A.
,
Zhang
,
J.
,
Zhang
,
Z.
,
Miller
,
W.
et al. 
(
1997
)
Gapped BLAST and PSI-BLAST: a new generation of protein database search programs
.
Nucleic Acids Res.
25
,
3389
3402
60
Mittelstädt
,
G.
,
Jiao
,
W.
,
Livingstone
,
E.K.
,
Moggré
,
G.J.
,
Nazmi
,
A.R.
and
Parker
,
E.J.
(
2018
)
A dimeric catalytic core relates the short and long forms of ATP-phosphoribosyltransferase
.
Biochem. J.
475
,
247
260
61
Krissinel
,
E.
(
2015
)
Stock-based detection of protein oligomeric states in jsPISA
.
Nucleic Acids Res.
43
,
W314
W319
62
Krissinel
,
E.
and
Henrick
,
K.
(
2004
)
Secondary-structure matching (SSM), a new tool for fast protein structure alignment in three dimensions
.
Acta Crystallogr., Sect. D: Biol. Crystallogr.
60
,
2256
2268
63
Berman
,
H.M.
,
Westbrook
,
J.
,
Feng
,
Z.
,
Gilliland
,
G.
,
Bhat
,
T.N.
,
Weissig
,
H.
et al. 
(
2000
)
The protein data bank
.
Nucleic Acids Res.
28
,
235
242
64
Sinha
,
S.C.
and
Smith
,
J.L.
(
2001
)
The PRT protein family
.
Curr. Opin. Struct. Biol.
11
,
733
739
65
Pisco
,
J.P.
,
de Chiara
,
C.
,
Pacholarz
,
K.J.
,
Garza-Garcia
,
A.
,
Ogrodowicz
,
R.W.
,
Walker
,
P.A.
et al. 
(
2017
)
Uncoupling conformational states from activity in an allosteric enzyme
.
Nat. Commun.
8
,
203
66
Motlagh
,
H.N.
,
Wrabl
,
J.O.
,
Li
,
J.
and
Hilser
,
V.J.
(
2014
)
The ensemble nature of allostery
.
Nature
508
,
331
339
67
Reid
,
K.S.C.
,
Lindley
,
P.F.
and
Thornton
,
J.M.
(
1985
)
Sulphur-aromatic interactions in proteins
.
FEBS Lett.
190
,
209
213
68
Fernando
,
N.
,
Manalil
,
S.
,
Florentine
,
S.K.
,
Chauhan
,
B.S.
and
Seneweera
,
S.
(
2016
)
Glyphosate resistance of C3 and C4 weeds under rising atmospheric CO2
.
Front. Plant Sci.
7
,
910
69
Kulis-Horn
,
R.K.
,
Persicke
,
M.
and
Kalinowski
,
J.
(
2015
)
Corynebacterium glutamicum ATP-phosphoribosyl transferases suitable for L-histidine production--Strategies for the elimination of feedback inhibition
.
J. Biotechnol.
206
,
26
37
70
Ufaz
,
S.
and
Galili
,
G.
(
2008
)
Improving the content of essential amino acids in crop plants: goals and opportunities
.
Plant Physiol.
147
,
954
961
71
Wycisk
,
K.
,
Kim
,
E.J.
,
Schroeder
,
J.I.
and
Krämer
,
U.
(
2004
)
Enhancing the first enzymatic step in the histidine biosynthesis pathway increases the free histidine pool and nickel tolerance in Arabidopsis thaliana
.
FEBS Lett.
578
,
128
134
72
Diederichs
,
K.
and
Karplus
,
P.A.
(
1997
)
Improved R-factors for diffraction data analysis in macromolecular crystallography
.
Nat. Struct. Biol.
4
,
269
275