Accepted Manuscript Title: Molecular modeling and molecular dynamics simulation study of archaeal leucyl-tRNA synthetase in complex with different mischarged tRNA in editing conformation Authors: A.V. Rayevsky, M. Sharifi, M.A. Tukalo PII: DOI: Reference:
S1093-3263(17)30265-6 http://dx.doi.org/doi:10.1016/j.jmgm.2017.06.022 JMG 6953
To appear in:
Journal of Molecular Graphics and Modelling
Received date: Revised date: Accepted date:
7-4-2017 7-6-2017 23-6-2017
Please cite this article as: A.V.Rayevsky, M.Sharifi, M.A.Tukalo, Molecular modeling and molecular dynamics simulation study of archaeal leucyl-tRNA synthetase in complex with different mischarged tRNA in editing conformation, Journal of Molecular Graphics and Modellinghttp://dx.doi.org/10.1016/j.jmgm.2017.06.022 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Molecular modeling and molecular dynamics simulation study of archaeal leucyl-tRNA synthetase in complex with different mischarged tRNA in editing conformation A. V. Rayevsky1*, M. Sharifi2, 3 and M. A. Tukalo1* 1
Institute of Molecular Biology and Genetics, NAS of Ukraine, 150 Academician Zabolotny Str., Kyiv 03680 Ukraine 2
Medway School of Pharmacy, Universities of Kent and Greenwich, Kent ME4 4TB, UK. Present address: 3Division of Systems Biology, National Center for Toxicological Research, US Food and Drug Administration, Jefferson, AR 72079
Emails:
[email protected];
[email protected]
Graphical abstract
1
Highlights Application and combining of in silico methods predicted the pre-intermediate state geometry, preceding the hydrolysis process in CP1 domain. Based on comparison of structural features in archaeal/eukaryal types of LeuRSs distinct modes of water attack and substrate activation were suggested. Two different substrates (cognate norvaline and non-cognate isoleucine) were used to reconstruct all steric conditions required to achieve the initial state of reaction. A possible mechanism of the post-transfer editing reaction in archaeal LeuRS system
is proposed. Abstract Aminoacyl-tRNA synthetases (aaRSs) play important roles in maintaining the accuracy of protein synthesis. Some aaRSs accomplish this via editing mechanisms, among which leucyl-tRNA synthetase (LeuRS) edits non-cognate amino acid norvaline mainly by post-transfer editing. However, the molecular basis for this pathway for eukaryotic and archaeal LeuRS remain unclear. In this study, a complex of archaeal P. horikoshii LeuRS (PhLeuRS) with misacylated tRNALeu was modeled wherever tRNA’s acceptor stem was oriented directly into the editing site. To understand the distinctive features of organization we reconstructed a complex of PhLeuRS with tRNA and visualize post-transfer editing interactions mode by performing molecular dynamics (MD) simulation studies. To study molecular basis for substrate selectivity by PhLeuRS’s editing site we utilized MD simulation of the entire LeuRS complexes using a diverse charged form of tRNAs, namely norvalyl-tRNALeu and isoleucyl-tRNALeu. In general, the editing site organization of LeuRS from P.horikoshii has much in common with bacterial LeuRS. The MD simulation results revealed that the post-transfer editing substrate norvalyl-A76, binds more strongly than isoleucyl-A76. Moreover, the branched side chain of isoleucine prevents water molecules from being closer and hence the hydrolysis reaction slows significantly. To investigate a possible mechanism of the post-transfer editing reaction, by PhLeuRS we have determined that two water molecules (the attacking and assisting water molecules) are localized near the carbonyl group of the amino acid to be cleaved off. These water molecules approach the substrate from the opposite side to that observed for Thermus thermophilus LeuRS (TtLeuRS). Based on the results obtained, it was suggested that the post-transfer editing mechanism of PhLeuRS differs from that of prokaryotic TtLeuRS. Keywords Aminoacyl-tRNA synthetases; Molecular dynamics; Molecular dynamics simulation, Post-transfer editing; Leucyl-tRNA synthetase, Pyrococcus horikoshii, tRNA
Introduction The main goal of the translation process is delivery of a tRNA molecule charged with cognate amino acid to the ribosome, where tRNA should be paired to the corresponding trinucleotide codon of mRNA. In this way, the protein synthesis machinery gets the amino 2
acid (bound to the corresponding tRNA) and is able to supply energy (ester bonding between amino acid and tRNA) required for peptide bond formation. A two-stage activation process includes formation of aminoacyl-adenylate from certain amino acid and ATP molecules with subsequent transfer of this amino acid to a CCA′-end of a cognate tRNA. The process takes place along the aminoacylation site of the corresponding aminoacyl-tRNA synthetase [1]. As seen in the equations below, the mechanism first requires the synthetase to facilitate the condensation of the corresponding amino acid with ATP to form aminoacyl-adenylate, prior to an aminoacyl-tRNA formation. Amino acid + ATP → Aminoacyl-AMP + PPi Aminoacyl-AMP + tRNA ↔ Aminoacyl-tRNA + AMP
There are 20 aaRSs, one for each individual amino acid. Based on some common features these aaRSs are divided into two classes. Class I contains a pair of characteristic conserved regions, specifically the structural domain (known as the Rossmann fold domain) and the monomeric active site which is formed by its parallel β-strands. Due to the extended binding site in this class, mostly large and hydrophobic radicals (e.g. hydroxyl radicals) are embedded inside the binding pocket. On the other hand, the aaRSs from class II can be characterized by its active sites, formed by anti-parallel β-sheet and flanked by α-helices [2, 3]. The aaRSs of the second class are mostly dimeric or multimeric and contain at least three conserved regions. In some cases aaRSs are unable to establish a high level of specificity against structurally similar amino acids in the synthetic reaction alone. To improve the accuracy of aminoacylation, some aaRSs have evolved the hydrolytic editing activity that destroys incorrectly formed aminoacyl-adenylate (the pre-transfer editing) or mischarged tRNAs (the post-transfer editing) [4, 5]. A subfamily of class 1A aminoacyl-tRNA synthetases, leucylisoleucyl- and valyl-tRNA synthetases (LeuRS, IleRS and ValRS respectively), which catalyze the attachment of similar non-polar amino acids onto their cognate tRNAs, are particularly closely related and possibly evolved from a common ancestor [5]. It is known that for LeuRSs from T. thermophilus, E.coli, P. horikoshii and S. cerevisiae there are three ways to keep the accuracy of the process. There are tRNA-independent pre-transfer editing of aminoacyl-adenylate, tRNA-dependent pre-transfer editing, caused by tRNA-induced with conformational changes, and post-transfer editing of the aminoacyl-tRNA by editing domain. Biochemical and structural studies have established that pre-transfer editing of LeuRSs take place in the aminoacylation site of the catalytic domain (CP core), at the same time posttransfer editing occurs in a separate editing domain [6-11]. LeuRS from Pyrococcus horikoshii consists of 967 residues. LeuRS from T. thermophilus is slightly shorter and contains 878 residues. Figure 1 depicts a generalized modular structure of LeuRS and the protein structures of P. horikoshii LeuRS, T.thermophilus LeuRS and ValRS. In class IA enzymes, the hydrolytic editing domain is a large (about 200 amino acids) domain, which called the conserved connective polypeptide 1 (CP1) domain. The domain is inserted into a core via two β-strand linkers. Importantly, the editing domain of PhLeuRS has a distinct orientation relative to the enzyme core, as compared with those of IleRS and ValRS, as well as the bacterial LeuRS [9]. 3
To date, the majority of knowledge about editing of LeuRS was obtained from prokaryotic enzymes [6-8, 11-16]. The crystal structure of Thermus thermophilus LeuRS in complex with an analogue of the post-transfer editing substrate, 2’-(L-norvalyl) amino-2’deoxyadenosine (Nva2AA), for the first time revealed the details of editing substrate binding in leucine system [7]. Furthermore, a structural model of TtLeuRS in complex with an alternative post-transfer substrate was constructed and subjected to quantum mechanics/molecular mechanics and molecular dynamics (QM/MM MD) simulations [17]. This modeling approach suggests that editing by LeuRS operates as a novel hybrid ribozyme/protein catalyst that is driven by the 3’-OH group of tRNA, whereas the amino acid residues of the editing site are crucial in promoting the reaction. Unlike LeuRSs of eukaryotes and archaea, there is no structural information on the complex of tRNA with the enzyme in the editing conformation or on the complex of the enzyme with the post-transfer editing substrate. On the other hand, the eukaryal/archaeal LeuRSs subtypes are architecturally distinct from bacterial LeuRSs [9, 12]. Therefore, the structural data and the proposed editing mechanism obtained for bacterial systems cannot be directly applied for eukaryotic and archaeal LeuRSs, which in turn should be the object of separate study. Knowledge of the editing mechanisms and the structural bases of substrate selection by aaRSs are important both for understanding the fundamental aspects of the translation of genetic information and for some applied aspects. The aaRSs are interesting targets for the development of anti-infective drugs [18-22]. The drug development procedure should be based on full and clear understanding of the mechanism, which provides a correct functioning of the target enzyme. Another prospect is in vivo incorporation of non-canonical amino acids as an elegant way for the chemical diversification of antimicrobial peptides [23]. In this study, a complex of archaeal P. horikoshii LeuRS with tRNALeu was modeled wherever tRNA’s acceptor stem was oriented directly into the editing site. Further, we investigated the editing mechanism against substrate in the archaeal system. All hypotheses were generated based on crystallographic data of TtLeuRS with norvaline analogue (NvaA76) as a candidate substrate for the post-transfer editing mechanism. To investigate individual features of PhLeuRS an MD simulation of the entire LeuRS complexes using a diverse charged form of tRNAs, namely Nva-tRNALeu and Ile- tRNALeu, was performed. Finally, on the basis of the data obtained, a possible mechanism of the post-transfer editing reaction in archaeal LeuRS system is suggested which differs from that for prokaryotic enzyme. Material and methods FT-MAP and HADDOCK approaches To determine key residues in the editing site two techniques were used: a local alignment of 3D-spatial structures and computational solvent mapping using the FT-MAP server [24]. The first method allows to compare LeuRS and ValRS structures from T.thermophilus and P.horikoshii (PDB IDs: 1GAX, 2WFG, 2V0G and 1WZ2), while the second method obtained a complex of truncated CP1together with screened small organic probes. Further, in order to generate a post-transfer conformation of the tRNA acceptor stem 4
region, the Haddock server [25] was used. Before docking, a set of active residues were defined based on alignment of CP1 domains from 2WFG and 2V0G crystal structures. Two networks of ambiguous interaction restraints (AIRs) between residues were developed to identify hot-spot residues, including specific interaction of the 3′-terminal adenosine with the editing site (A76), in archaeal and bacterial synthetases. Both semi- and fully-flexible methods were used to provide a hybrid docking procedure. Larger part of tRNA stem bound to a catalytic core and the CP1 domain were set as rigid with flexible patterns (seven nucleotides) while both terminal regions of tRNA and the linker between the CP1 and catalytic core were docked in flexible mode. Docking procedure Docking of two substrate analogs (Nva-A76/Ile-A76) was conducted in AutoDock 4.2 [26]. Hydrogen atoms were added to ligands and proteins using Python script collection in AutoDock Tools (http://autodock.scripps.edu/). The Gasteiger partial charges were assigned to atoms of ligand molecules and all grids of interaction energies were calculated using AutoGrid map and AMBER force field based on the macromolecular target. The cubic grid box of 55 Å size (x, y, z) with a spacing of 0.375 Å and grid map covered the LeuRS editing site based on FT-MAP server results. The Lamarckian genetic algorithm (LGA) was utilized for enzyme-ligand docking with population size of 150 and with a maximum of 25,000 generations and a crossover rate of 0.8 and 10 docking runs. A ligand-free CP1 domain of LeuRS (PDB ID: 1WZ2) was set to rigid while a group of residues (namely, Thr229, Asp332, Asp314 and Asn317) in the editing site and all ligands were treated as flexible. A 3D model was used as a rigid scaffold and to explore different conformers in the active site, both ligand and residues in the editing site were set to the flexible mode. All figures of protein-tRNA interactions were prepared with visual molecular dynamics (VMD) software [27]. Correspondingly, for the ligand docking, four various points of torsion and a structure-validation web service (MolProbity) was used for structural evaluation of interatomic contacts in the docked regions [28].
Molecular dynamics simulation All the MD simulations were performed in GROMACS, a highly parallel open source molecular simulation toolkit (version 4.0.5) [29] generalized with the Amber99 force field. A Modeler v.8 [30] loop optimization method was used to rebuild and refine a set of missing residues of CP1 domain. RED III (http://upjv.q4md-forcefieldtools.org/RED/) software was used to prepare post-transfer substrates (aa-tRNA) and then to generate a topology of a fulllength tRNA molecule, a basic preparation tool LEaP module (Amber package, version 7) was applied [31]. Based on the MD simulation protocol used, each system was solvated in a water box (in the TIP3P water models) in the range of 12 Å were considered. The entire systems were neutralized by adding Na+ counter ions and replacing solvent molecules. In the initial phase of the MD simulation (relaxation), the water molecules were relaxed at 310 K for 25 ps, while the position of atoms in the protein and tRNALeu were adjusted by a harmonic biasing function using a force constant of 500 kcal/mol per A2. Further, in order to have an accurate 5
transition from steered MD to the uncustomized MD, the force constant was reduced to 250, 125, 50, 25, 10 and 5 kcal/mol per A2 in six MD simulations each of 500ps. The entire protein-tRNA complex was relaxed during 50 ns and relative distance constraints between A76 and Asp332 was applied. Finally, last 10 ns of trajectory were used to cluster CP1 domain conformations in 10 clusters for ensemble docking and All six simulations performed under periodic boundary conditions in MD and thermalized in NPT ensemble by using Berendsen's coupling algorithm (to maintain the constant temperature of 333 K, physiologically normal temperature for thermophilic proteins, and under the pressure constant of 1 atm bar). The electrostatic interactions were calculated by using the Particlemesh Ewald (PME) algorithm, with an interpolation order of 4 and a grid spacing of 0.12 nm. The van der Waals forces were set to a cutoff of 10 Å and the coordinates were stored every 100 ps. All the analyses of the MD simulations were carried out using GROMACS and the computations were performed in a high performance Linux cluster computer systems [32]. MolProbity server was used to evaluate the protein qualities obtained from MD simulation. All superposition states were generated with Chimera package [http://www.cgl.ucsf.edu/chimera], and then structures were visualized in Discovery studio 3.0.
Results and Discussion Modeling of PhLeuRS-tRNALeu in post-transfer conformation Due to the lack of X-ray crystallographic structures of P.horikoshii LeuRS with cocrystallized post-transfer ligands and LeuRS-tRNA complex in post-transfer conformation, it was necessary to first determine which amino acids residues are involved in the binding of substrate, moreover, it assists to calculate a probable localization of the tRNA’s 3'-end oriented in the editing site. To build a model of archaeal LeuRS in the post-transfer editing conformation, initially two available structures were compared with other aaRSs from wellstudied species. In this way, several common features were identified in the modular structure as well as a separate CP1 (editing) domain. This information was used to determine a group of residues which were involved in the process of substrate recognition. BLAST (Basic Local Alignment Tool) search for resemble structures with a ligand in the editing site was performed to identify existing homologues through the UniProt [33] database. In the end the CP1 domain from P.horikoshii LeuRS (PDB ID: 1WZ2) has the following values of similarity as 36%, 31% and 31% against CP1 domains from H. sapiens LeuRS (PDB ID: 2WFD), T.thermophilus ValRS (PDB ID: 1GAX) and C.albicans LeuRS (PDB ID: 2WFG), respectively. Despite the low degree of the sequence similarity obtained, the figure 2 presents a satisfactory 3D alignment and comparison of CP1 domains from ValRS, and TtLeuRS. FT-MAP server identified numerous “hot-spot” binding regions on the surface of the P.horikoshii enzyme. This algorithm performs a global search of the entire protein surface 6
for regions that bind a number of small organic probe molecules (e.g. phenol, acetone, benzene, and cyclohexane). A truncated CP1 domain from P.horikoshii LeuRS was processed with the FT-MAP server and the small probes located near Thr229, Asp332, Asp314 and Asn317 and then a crystal structure of C.albicans LeuRS (with a benzoxaborole-AMP adduct bound in the editing domain) was aligned to the archaeal structure with probes. Thus the location of probes and ligand matched (Figure 2B). Finally all the residues involved in ligand binding were selected after FT-MAP probing of archaeal domain and were compared accordingly. The results of cross-check analysis showed that the mechanism of selectivity and fixation of substrate by PhLeuRS were similar to those demonstrated with TtLeuRS. Further, molecular docking results of LeuRS CP1 domain with Nva-A76 and Ile-A76 were applied to build initial structures for MD simulation studies. Two complexes with best scored poses of substrates, which adenosine part and amino group formed interactions with Asp314, Asn317 and Asp332 amino acids, were compared with crystal structures of the CP1 domain from T.thermophilus ValRS (PDB ID: 1GAX) and C.albicans LeuRS (PDB ID: 2WFG). This alignment showed the same location of purine bases and orientation of amino groups of substrates. After identification of important residues (those with a role in fixation of A76), a flexible docking of tRNA molecule into synthetase was performed in the HADDOCK server, and the two best conformations (Figure 3) from a set of generated clusters were selected primarily by RMSD, AIRs violations, intermolecular hydrogen bonds analysis and analysis of hydrophobic contacts. Both conformations of the stem were optimized in Gromacs package in order to equilibrate in MD simulations. For this purpose, initially a steer MD mode was used to stabilize the complexes during the first run of simulation (3 ns) and the purine base of A76 was fixed in the site with distance constraints through COM (Center of Mass) velocity model to restrain unfavorable movement of the 3'-CCA terminal of tRNA. In this way, the most stable position of A76 was found in the site which then was attached to the Asp314 and the Asn317 carboxyl chain atoms by geometry constraints. After a steer molecular dynamics (3 ns), to start a free MD simulation, the force constraint strategy (2000 kJ/mol per A2) has been eliminated further. Both PhLeuRS trajectories with norvalyl- and isoleucyl-tRNA were analyzed and then most frequent (in terms of stability) structures were clustered into 10 groups to be used further in post-transfer editing mechanism evaluation.
Post-transfer substrates binding and possible hydrolysis mechanism of misacylated tRNA Two different MD simulations were carried out to demonstrate the binding properties of post-transfer substrates (Nva-A76/Ile-A76). A complex of full-length LeuRS with tRNALeu was previously equilibrated during 50 ns of free MD simulations. Conformations were clustered and six ligand docking poses accordingly were selected, and the best poses in complex with LeuRS were used for a set of 5 ns calculations. As can be seen in Figure 3, the 3'-end stem of tRNA forms several strong hydrogen bonds with a linker part of LeuRS and 7
the CP1 domain. The stability of MD simulation was tested and validated by backbone RMSD and by comparison of total energies. The RMSD fluctuations were stabilized for Nva-A76 soon after 2000 ps, as opposed to the Ile-A76 compound (Figure 4A). To explore a binding strength of post-transfer substrates, the number of intermolecular hydrogen bonds was analyzed and showed the average numbers of 7.3 and 5.1 between NvaA76/ Ile-A76 and LeuRS. The most stable bonds were formed with Met323, Asp332 and Lys398. The coulombic interaction value between Nva-A76 and the CP1 domain were stable throughout the 5 ns simulation with an average of -352.1 kJ/mol, while for Ile-A76, a level of coulombic interaction energy with an average of -175.3 kJ/mol (Fig. 4B) was observed. The graphs in Figure 4 demonstrate the differences between the rates of electrostatic interaction energies (kJ/mol) (Fig. 4B) and ligand RMSDs in Nva-A76 and Ile-A76 (Fig. 4A). For hydrolysis of the ester bond between the mischarged amino acid and tRNA a water molecule must be activated by a base for nucleophilic attack of the carbonyl carbon of the scissile ester. To identify the possible attacking water molecule all ligand-protein complexes with water-mediated contacts were obtained. It is worth mentioning that only those stable water molecules (including those that were not disposed during more than 20 ps) were considered to be included in the complex system. Further, it was shown that the number of waters near the Nva-A76 during the 5ns MD was 10-fold higher than those near the Ile-A76. Aminoacyl-tRNA complex (Nva-A76 /Ile-A76) as ester together with surrounding water molecules were tested to estimate the frequency of the state, when the appropriate geometry between oxygen of the water and carboxyl group has reached. The most likely theory for the nucleophile attack confirms that the water molecule can attack the carboxyl carbon after activation by atoms of nearest residues or 3'-OH of A76 tRNA. One of the MD limitations of this reaction includes the distance amplitude. Among all analyzed water molecules (around the catalytic site) the one found to be closest to the carbonyl carbon of the norvaline. The atomic distance of 3.1 Å between water oxygen and carbonyl atom of aminoacyl allows to suggest, that this water molecule acts as a nucleophile in the post-transfer editing reaction. Other limitations refer to the orientation of water molecule relative to the plane of carboxyl group. In this way, atoms of amino group, carboxyl group and C-alpha can form a plane and the favorable angle generated by the water molecule is estimated to be 103° (Fig. 5A). Further limitations comprises activation of water molecules (e.g. with Asp332 Thr234 or 3'-OH of A76 tRNA) and the stability of the water molecules during MD simulation to provide enough time for the reaction. In the case of NvaA76 the water molecule W1, which took the position of attack formed a H-bond with the nearest water molecule W2 (the assisting water molecules), which could be activated by the conserved Thr234 (Fig. 5B). Alternatively, the Ile-A76 molecule can also be attacked by a water molecule. The results of our MD simulations have shown that Ile-A76 occupied the same place and orientation as Nva-A76, but the branched side chain prevents water molecules from being closer and hence the hydrolysis reaction slows significantly. Interestingly, the alignment of 8
averaged structures during the last 2 ns of both dynamics disclosed similar disposition of Nva-A76 and Ile-A76 (Fig. 5C). Nevertheless, in Ile-A76, a methyl group protrudes out of the binding pocket causing a steric obstacle for solvent accessibility. All water molecules near Ile-A76 did not form any interaction with the protein or other water molecules. In the case of PhLeuRS, the attacking water molecule approaches the substrate from the Met323 side, while in the case of TtLeuRS [7, 17] the water molecule approaches the opposite side, Asp344 (Ala329 in the PhLeuRS). Perhaps this is due to some differences in the structure of the active centers of the editing domains of prokaryotic and eukaryal/archaeal enzymes. This in turn is caused by some differences in the primary structure of both enzymes in the region of the editing active site. For example, the residue Thr248 in TtLeuRS, conserved in prokaryotic LeuRSs, is replaced by Leu230, and Asp344 in TtLeuRS is replaced by Ala or Ser in LeuRSs from archaea and eukaryote.
Conclusions LeuRS is unique among aaRSs with post-transfer editing activity, since the activity of its editing domain is primarily aimed to prevent mis-incorporation of the non-protein amino acid norvaline [16]. The editing domain has been proposed to be an ancient addition to the aminoacylation core of IleRS, ValRS and LeuRS [5]. Structural and functional aspects of proofreading mechanism of class 1A prokaryotic aaRSs have been extensively studied, and a mechanism of hydrolysis of misacylated tRNA has already been proposed for bacterial TtLeuRS [4, 5, 7, 13-15 and 17]. The eukaryotic and archaeal aaRSs of this subclass have not been studied to the same extent. In this study, a docking model of P. horikoshii LeuRS with misacylated tRNALeu was generated wherever tRNA’s acceptor stem was oriented into the editing site. The analysis of PhLeuRS structure revealed common domain organization with a group of LeuRS, ValRS and IleRS from T.thermophilus. Moreover, the CP1 domain consists of similar structural elements to those identified in LeuRSs from C.albicans and H.sapiens. In general, the editing site organization of LeuRS from P.horikoshii has much in common with bacterial LeuRS. Based on these conclusions docking poses of post-transfer editing substrates Nva-76 and Ile-76 in the editing site of PhLeuRS were obtained. Despite the difference in interaction between tRNALeu and enzymes from T.thermophilus and P.horikoshii, the post-transfer editing substrate binding mode is quite similar. As in the case of bacterial TtLeuRS [7], the α-amino group of the norvaline forms H-bonds with the carboxyl group of Asp332 (Asp347 in the TtLeuRS). The next important anchor of the substrate in TtLeuRS is Thr247 and Thr248, the conservative elements of the threonine rich peptide of the bacterial enzymes. Due to the fact that in the archaeal LeuRSs the Thr248 is replaced by Leu230, this role in our model is performed by Glu394, which makes a hydrogen bond to the 3’OH of the ribose. To investigate a possible mechanism of the reaction of PhLeuRS post-transfer editing the water molecules of interest were localized near the carbonyl group of amino acid to be cleaved off. The remained water molecules were evaluated for the position of hydrogen atoms and angle of atom approximation and MD frames and two molecules were selected. 9
One of them (W1), the attacking water molecule, is poised to perform an attack at the electrophilic carbonyl group to hydrolyze the ester bond. The attacking water molecule approaches the substrate from the opposite side than in the case of TtLeuRS [17]. This water molecule forms H-bonds with the second, the assisting H2O molecule (W2), which, in turn, forms H-bond with a Thr234 residue. The second water molecule was supposed to participate in the process and further facilitate the hydrolysis. The participation of two water molecules in the hydrolysis of the erroneous product was previously proposed for the mechanism of post-transfer editing by prolyl-tRNA synthetase [34]. The post-transfer editing mechanism by PhLeuRS differs from that for prokaryotic TtLeuRS, the editing active site of which has been suggested to operate using a novel hybrid ribozyme/protein catalyst to clear mischarged tRNALeu. In the model for TtLeuRS, the 3’-OH group of the terminal A76 of tRNALeu has been proposed to activate a nucleophilic water molecule and hence drive ester bond cleavage [17]. Alternatively, in the model for PhLeuRS the 3’-OH group of the terminal A76 forms H-bond with Glu394 and is far away from the attacking water molecule. To study molecular basis for substrate selectivity by the PhLeuRS’s editing site we utilized MD simulation of the entire LeuRS complex using a diverse charged form of tRNAs, namely norvalyl-tRNALeu and isoleucyl-tRNALeu. The results of MD simulations revealed that the post-transfer editing substrate norvalyl-A76 binds more strongly than isoleucyl-A76. A suggested above mechanism of hydrolysis corresponds to the situation with post-transfer editing of Nva-A76, which is considered to be the main (relevant) post-transfer substrate for LeuRS. However, Ile-A76 binds less strongly to the editing site and is less accessible to water the molecules that make this substrate less suitable. Interestingly, for bacterial LeuRS from E. coli, it was experimentally shown that the aminoacyl-tRNA hydrolysis rate measured by the single-turnover method for Nva-tRNA proceeds an order of magnitude faster than for Ile-tRNA [15]. Declaration of interest The authors have declared that no conflict of interest exists. The views presented in this article are those of the author and do not necessarily reflect those of the US Food and Drug Administration. No official endorsement is intended nor should be inferred.
10
References: 1. Ibba, M. and Soll, D. (2000) Aminoacyl-tRNA synthesis. Annu. Rev. Biochem. 69, 617-650. 2. Cusack, S., Berthet-Colominas, C., Hartlein, M., Nassar, N. and Leberman, R. (1990) A second class of synthetase structure revealed by X-ray analysis of Escherichia coli seryl-tRNA synthetase at 2.5 Å resolution. Nature 347, 249-255. 3. Eriani, G., Delarue, M., Poch, O., Gangloff, J. and Moras, D. (1990) Partition of tRNA synthetases into two classes based on mutually exclusive sets of sequence motifs. Nature 347, 203-206. 4. Fersht, A. R. (1977) Editing mechanisms in protein synthesis. Rejection of valine by the isoleucyl-tRNA synthetase. Biochemistry 16, 1025-1030. 5. Yadavalli, S.S. and Ibba, M. (2012) Quality control in aminoacyl-tRNA synthesis: its role in translational fidelity. Adv. Protein Chem. Struct. Biol. 86, 1-43. 6. English, S., English, U., von der Haar, F. and Cramer, F. (1986) The proofreading of hydroxyl analogues of leucine and isoleucine by leucyl-tRNA synthetases from E. coli and yeast. Nucleic Acids Res. 14, 7529-7539. 7. Lincecum, T. L., Tukalo, M., Yaremchuk, A., Mursinna, R. S., Williams, A. M., Sproat, B. S., Van Den Eynde, W., Link, A., Van Calenbergh, S., Grøtli, M., Martinis, S. A. and Cusack, S. (2003) Structural and mechanistic basis of pre- and posttransfer editing by leucyl-tRNA synthetase. Mol. Cell. 11, 951– 963. 8. Boniecki, M. T., Vu, M.T., Betha, A. K. and Martinis, S. A. (2008) CP1-dependent partitioning of pretransfer and posttransfer editing in leucyl-tRNA synthetase. Proc. Nat. Acad. Sci. USA 105, 19223– 19228. 9. Fukunaga, R. and Yokoyama, S. (2005) Crystal structure of leucyl-tRNA synthetase from the archaeon Pyrococcus horikoshii reveals a novel editing domain orientation. J. Mol. Biol. 346, 57-71. 10. Sarkar, J. and Martinis, S.A. (2011) Amino-acid-dependent shift in tRNA synthetase editing mechanisms. J. Am. Chem. Soc. 133, 18510-18513. 11. Perona, J.J. and Gruic-Sovulj, I. (2013) Synthetic and editing mechanisms of aminoacyl-tRNA synthetases. Top. Curr. Chem. 344, 1-41. 12. Cusack, S., Yaremchuk, A. and Tukalo, M. (2000) The 2 Å crystal structure of leucyl-tRNA synthetase and its complex with a leucyl-adenylate analogue. EMBO J. 19, 2351-2361. 13. Tukalo, M., Yaremchuk, A., Fukunaga, R., Yokotama, S. and Cusack, S. (2005) The crystal structure of leucyl-tRNA synthetase complexed with tRNALeu in the post-transfer-editing conformation. Nat. Struct. Mol. Biol. 12, 923-930. 14. Palencia, A., Crépin, T., Vu, M. T., Lincecum, T. L., Martinis, S. A. and Cusack, S. (2012) Structural dynamics of the aminoacylation and proofreading functional cycle of bacterial leucyl-tRNA synthetase. Nat. Struct. Mol. Biol. 19, 677-685. 15. Cvetesic, N., Perona, J.J. and Gruic-Sovulj, I. (2012) Kinetic partitioning between synthetic and editing pathways in class I aminoacyl-tRNA synthetases occurs at both pre-transfer and post-transfer hydrolytic steps. J. Biol. Chem. 287, 25381-25394. 16. Cvetesic, N., Palencia, A., Halasz, I., Cusack, S. and Gruic-Sovulj, I. (2014) The physiological target for LeuRS translational quality control is norvaline. EMBO J. 33, 1639–1653. 17. Hagiwara, Y., Field, M. J., Nureki, O. and Tateno, M. (2010) Editing mechanism of aminoacyl-tRNA synthetase operates by a hybrid ribozyme/protein catalyst. J. Am. Chem. Soc. 132, 2751–2758.
11
18. Rock, F.L., Mao, W., Yaremchuk, A., Tukalo, M., Crepin, T., Zhou, H., Zhang,Y.-K., Hernandez, V., Akama, T., Baker, S.J., Plattner, J.J., Shapiro, L., Martinis, S.A., Benkovic, S.J., Cusack, S. and Alley, M.R.K. (2007) An antifungal agent inhibits an aminoacyl-tRNA synthetase by trapping tRNA in the editing site. Science 316, 1759-1761. 19. Lv, P.C. and Zhu, H.L. (2012) Aminoacyl-tRNA synthetase inhibitors as potent antibacterials. Curr. Med. Chem. 19, 3550-3563. 20. Jain, V., Yogavel, M., Oshima, Y., Kikuchi, H., Touquet, B., Hakimi, M.A. and Sharma, A. (2015) Structure of prolyl-tRNA synthetase-halofuginone complex provides basis for development of drugs against malaria and toxoplasmosis. Structure 23, 819-829. 21. Gudzera, O.I., Golub, A.G., Bdzhola, V.G., Volynets, G.P., Lukashov, S.S., Kovalenko, O.P., Kriklivyi, I.A., Yaremchuk, A.D., Starosyla, S.A., Yarmoluk, S.M. and Tukalo, M.A. (2016) Discovery of potent anti-tuberculosis agents targeting leucyl-tRNA synthetase. Bioorg. Med. Chem. 24, 1023-1031. 22. Saint-Léger, A., Sinadinos, C. and Ribas de Pouplana, L. (2016) The growing pipeline of natural aminoacyl-tRNA synthetase inhibitors for malaria treatment. Bioengineered. 7, 60-64. 23. Baumann, T., Nickling, J.H., Bartholomae, M., Buivydas, A., Kuipers, O.P., Budisa, N. (2017) Prospects of In vivo Incorporation of non-canonical amino acids for the chemical diversification of antimicrobial peptides. Front Microbiol. 8, 1-9. 24. Ngan, C. H., Bohnuud, T., Mottarella, S. E., Beglov, D., Villar, E.A., Hall, D.R., Kozakov, D. and Vajda, S. (2012) FTMAP: extended protein mapping with user-selected probe molecules. Nucleic Acids Res. 40, W271–275. 25. Dominguez, C., Boelens, R. and Bonvin, A.M.J.J. (2003) HADDOCK: a protein-protein docking approach based on biochemical and/or biophysical information. J. Am. Chem. Soc. 125, 1731-1737. 26. Morris, G. M., Huey, R., Lindstrom, W. and Sanner, M. F. (2009) Autodock4 and AutoDockTools4: automated docking with selective receptor flexibility. J. Computational Chemistry 16, 2785-2791. 27. Humphrey, W., Dalke, A. and Schulten, K. (1996) VMD-Visual Molecular Dynamics. J. Molec. Graphics 14, 33-38. 28. Davis, I.W., Leaver-Fay, A., Chen, V.B., Block, J.N., Kapral, G.J., Wang, X., Murray, L.W., Arendall, W.B. 3rd, Snoeyink, J., Richardson, J.S., and Richardson, D.C. (2007) MolProbity: all-atom contacts and structure validation for proteins and nucleic acids. Nucleic Acids Res. 35, W375-383. 29. Hess, B., Kutzner, C., van der Spoel, D. and Lindahl, E. (2008) GROMACS 4: Algorithms for highly efficient, load-balanced, and scalable molecular simulation. J. Chem. Theory Comput. 4, 435-447. 30. Webb, B. and Sali, A. (2014) Protein structure modeling with MODELLER. Methods Mol Biol. 1137, 115. 31. Case, D.A., Pearlman, D.A., Caldwell, J.W., Cheatham III, T.E., Wang, J., Ross, W.S., Simmerling, C.L., Darden, T.A., Merz, K.M., Stanton, R.V., Cheng, A.L., Vincent, J.J., Crowley, M., Tsui, V., Gohlke, H., Radmer, R.J., Duan, Y., Pitera, J., Massova, I., Seibel, G.L., Singh, U.C., Weiner, P.K. and Kollman P.A. (2002) AMBER 7, University of California, San Francisco. 32. Salnikov, A., Sliusar, I., Sudakov, O., Savytskyi, O. and Kornelyuk, A. (2010) Virtual laboratory moldyngrid as a part of scientific infrastructure for biomolecular simulations. International J. Computing 9, 294-300. 33. The UniProt Consortium (2017) UniProt: the universal protein knowledgebase Nucleic Acids Res. 45, D158-D169. 34. Boyarshin, K., Priss, A., Rayevskiy, A., Ilchenko, M., Dubey, I., Kriklivyi, I., Yaremchuk, A. and Tukalo, M. (2017) A new mechanism of post-transfer editing by aminoacyl-tRNA synthetases: catalysis of hydrolytic reaction by bacterial-type prolyl-tRNA synthetase. J. Biomol. Struct. Dynam. 35, 669-682.
12
Figure 1. The modular structure of LeuRS from P.horikoshii, LeuRS and ValRS from T.thermophilus. The Rossmann-fold domain (orange), the connective peptide 1 (CP1 editing domain) (cyan), the CP1 hairpin (blue), the connective peptide 2 (CP2 domain) (green), the CP core (white), the stem contact (SC-fold domain) (red), the α-helix bundle domain (magenta) and the Leucyl-specific domains 1 are shown in yellow. General representation of the protein sequence with mapped functional domains shown as a bar at the bottom.
Figure 2. The figure demonstrates the comparative alignment of LeuRS structures. A) The LeuRS structures from P.horikoshii (red) against LeuRS (green) from T.thermophilus. Aminoacylation and editing sites are represented with spheres of the corresponding color to mark active site location. A dislocation of aminoacylation sites in catalytic cores matches well, despite the difference in size of the cavity. However, the spatial location of the editing site matched only in the pair of PhLeuRS and TtValRS. Both CP1 editing domains of LeuRS from P.horikoshii (red) and T.thermophilus (green) have the same twist of the linker part and, as a result, the same orientation relative to the catalytic core with a large distance (25Å) 13
between both editing sites. B) Comparison of the FT-MAP probing (orange lines) of the CP1 from P.horikoshii (red) and cocrystallized ligand in the editing site of C.albicans (cyan).
Figure 3. Two best scored orientations of docked 3'-terminal stem of tRNA directed to the editing site. A) Figure demonstrates a docked conformation of tRNA, which interacts with the protein with strong hydrogen bonds. B) Different conformation of the same pattern (nucleotides from 73-80) of the tRNA chain from two docking poses. After MD, the more stable (marked with green) conformation was used in the study. Another conformation (marked with cyan) was deformed during MD and most important interactions with the protein. The critically deformed region is shown with black arrow.
Figure. 4. The RMSD of ligand structure (A) and short range columbic interaction energy with protein environment (B).
14
Figure 5. The suggested mechanism of nucleophile attack at the carbonyl carbon of Nva-A76 by water molecule (W1) and formation of hydrogen bonds with surrounding amino acids (A) and demonstration of chain water mediated activation by Thr234 from another view (B). The 3D alignment of both substrates Nva-A76 (green) and Ile-A76 (cyan), the averaged structures were generated from the last 5 ns of MD (C).
Figure 5. The suggested mechanism of nucleophile attack at the carbonyl carbon of Nva-A76 by water molecule (W1) and formation of hydrogen bonds with surrounding amino acids (A) and demonstration of chain water mediated activation by Thr234 from another view (B). The 3D alignment of both substrates NvaA76 (green) and Ile-A76 (cyan), the averaged structures were generated from the last 5 ns of MD (C).
15
