Advertisement

Discovery of two distinct aminoacyl-tRNA synthetase complexes anchored to the Plasmodium surface tRNA import protein

Open AccessPublished:April 26, 2022DOI:https://doi.org/10.1016/j.jbc.2022.101987
      Aminoacyl-tRNA synthetases (aaRSs) attach amino acids to their cognate transfer RNAs. In eukaryotes, a subset of cytosolic aaRSs is organized into a multisynthetase complex (MSC), along with specialized scaffolding proteins referred to as aaRS-interacting multifunctional proteins (AIMPs). In Plasmodium, the causative agent of malaria, the tRNA import protein (tRip), is a membrane protein that participates in tRNA trafficking; we show that tRip also functions as an AIMP. We identified three aaRSs, the glutamyl-tRNA synthetase (ERS), glutaminyl-tRNA synthetase (QRS), and methionyl-tRNA synthetase (MRS), which were specifically coimmunoprecipitated with tRip in Plasmodium berghei blood stage parasites. All four proteins contain an N-terminal glutathione-S-transferase (GST)–like domain that was demonstrated to be involved in MSC assembly. In contrast to previous studies, further dissection of GST-like interactions identified two exclusive heterotrimeric complexes: the Q-complex (tRip–ERS–QRS) and the M-complex (tRip–ERS–MRS). Gel filtration and light scattering suggest a 2:2:2 stoichiometry for both complexes but with distinct biophysical properties and mutational analysis further revealed that the GST-like domains of QRS and MRS use different strategies to bind ERS. Taken together, our results demonstrate that neither the singular homodimerization of tRip nor its localization in the parasite plasma membrane prevents the formation of MSCs in Plasmodium. Besides, the extracellular localization of the tRNA-binding module of tRip is compensated by the presence of additional tRNA-binding modules fused to MRS and QRS, providing each MSC with two spatially distinct functions: aminoacylation of intraparasitic tRNAs and binding of extracellular tRNAs. This unique host–pathogen interaction is discussed.

      Keywords

      Abbreviations:

      β-Me (2-mercaptoethanol), aaRS (aminoacyl-tRNA synthetase), AIMP (aaRS-interacting multifunctional protein), CDD (conserved domain database), coIP (coimmunoprecipitation), DDM (β-D-maltoside), DLS (dynamic light scattering), EMAPII (endothelial monocyte-activating polypeptide II), ERS (glutamyl-tRNA synthetase), EPRS (glutamyl–prolyl-tRNA synthetase), LCR (low-complexity region), MALS (multiangle light scattering), MRS (methionyl-tRNA synthetase), MS (mass spectrometry), MSA (multisequence alignment), MSC (multi-aaRS complex), MW (molecular weight), Ni-NTA (nickel-nitrilotriacetic acid), PDB (Protein Data Bank), QRS (glutaminyl-tRNA synthetase), SEC (size-exclusion chromatography), SLS (static light scattering), tRip (tRNA import protein)
      Aminoacyl-tRNA synthetases (aaRSs) are a family of essential enzymes that perform the first step in protein synthesis by specifically attaching an amino acid to its corresponding tRNA (
      • Rubio Gomez M.A.
      • Ibba M.
      Aminoacyl-tRNA synthetases.
      ). In addition to their canonical role in tRNA aminoacylation, eukaryotic aaRSs have evolved to participate in a wide range of alternative functions (
      • Park S.G.
      • Ewalt K.L.
      • Kim S.
      Functional expansion of aminoacyl-tRNA synthetases and their interacting factors: New perspectives on housekeepers.
      ,
      • Pang Y.L.J.
      • Poruri K.
      • Martinis S.A.
      tRNA synthetase: tRNA aminoacylation and beyond.
      ). These functions include regulation of gene expression, RNA processing and trafficking, as well as cell signaling activities (
      • Guo M.
      • Schimmel P.
      Essential nontranslational functions of tRNA synthetases.
      ,
      • Hyeon D.Y.
      • Kim J.H.
      • Ahn T.J.
      • Cho Y.
      • Hwang D.
      • Kim S.
      Evolution of the multi-tRNA synthetase complex and its role in cancer.
      ). In eukaryotes, the association of several aaRSs within a multi-aaRS complex (MSC) partitions the translational activities of aaRSs while in the MSC, from their alternative functions once dissociated from the MSC (
      • Han J.M.
      • Kim J.Y.
      • Kim S.
      Molecular network and functional implications of macromolecular tRNA synthetase complex.
      ,
      • Lee S.W.
      • Cho B.H.
      • Park S.G.
      • Kim S.
      Aminoacyl-tRNA synthetase complexes: Beyond translation.
      ,
      • Cui H.
      • Kapur M.
      • Diedrich J.K.
      • Yates J.R.
      • Ackerman S.L.
      • Schimmel P.
      Regulation of ex-translational activities is the primary function of the multi-tRNA synthetase complex.
      ). In addition to a subset of cytosolic aaRSs, MSCs contain up to three accessory proteins referred as aaRS-interacting multifunctional proteins (AIMPs) (Fig. 1A) (reviewed in (
      • Guo M.
      • Yang X.-L.
      Architecture and metamorphosis. In aminoacyl-tRNA synthetases in biology and medicine.
      ,
      • Havrylenko S.
      • Mirande M.
      Aminoacyl-tRNA synthetase complexes in evolution.
      )). One AIMP is common to all known MSCs. In unicellular eukaryotes, it is a single polypeptide characterized by a specific structural organization: an N-terminal domain with homology to glutathione transferases (glutathione-S-transferase [GST]–like domain) fused through a poly-lysine linker to a C-terminal endothelial monocyte-activating polypeptide II (EMAPII)–like domain (
      • Karanasios E.
      • Simos G.
      Building arks for tRNA: Structure and function of the Arc1p family of non-catalytic tRNA-binding proteins.
      ,
      • Kapps D.
      • Cela M.
      • Théobald-Dietrich A.
      • Hendrickson T.
      • Frugier M.
      OB or Not OB: Idiosyncratic utilization of the tRNA-binding OB-fold domain in unicellular, pathogenic eukaryotes.
      ) (Fig. 1A). GST-like domains are known to mediate protein–protein interactions, while EMAPII-like domains have been shown to recognize the tRNA 3D structure. This AIMP was first identified in the yeast Saccharomyces cerevisiae (Arc1p (
      • Simos G.
      • Segref A.
      • Fasiolo F.
      • Hellmuth K.
      • Shevchenko A.
      • Mann M.
      • et al.
      The yeast protein Arc1p binds to tRNA and functions as a cofactor for the methionyl- and glutamyl-tRNA synthetases.
      )) and in pathogenic parasites like Toxoplasma gondii (Tg43 (
      • van Rooyen J.M.
      • Murat J.-B.
      • Hammoudi P.-M.
      • Kieffer-Jaquinod S.
      • Coute Y.
      • Sharma A.
      • et al.
      Assembly of the novel five-component apicomplexan multi-aminoacyl-tRNA synthetase complex is driven by the hybrid scaffold protein tg-p43.
      )) and Trypanosoma brucei (MCP1 (
      • Cestari I.
      • Kalidas S.
      • Monnerat S.
      • Anupama A.
      • Phillips M.A.
      • Stuart K.
      A multiple aminoacyl-tRNA synthetase complex that enhances tRNA-aminoacylation in african trypanosomes.
      )). In metazoans, it is assembled from two separate peptides, either AIMP1 (p43, EMAPII-like domain) and AIMP2 (p38, GST-like domain) in vertebrates (
      • Ahn H.-C.
      • Kim S.
      • Lee B.-J.
      Solution structure and p43 binding of the p38 leucine zipper motif: coiled-coil interactions mediate the association between p38 and p43.
      ) or the C-terminal domain (EMAPII-like domain) of the methionyl-tRNA synthetase (MRS) and AIMP2 in nematodes (
      • Havrylenko S.
      • Legouis R.
      • Negrutskii B.
      • Mirande M.
      Caenorhabditis elegans evolves a new architecture for the multi-aminoacyl-tRNA synthetase complex.
      ). The bifunctional organization of these AIMPs (tRNA binding and protein–protein interaction domains) is crucial for MSCs to participate in tRNA channeling (
      • Negrutskii B.S.
      • Deutscher M.P.
      Channeling of aminoacyl-tRNA for protein synthesis in vivo.
      ,
      • Simos G.
      • Sauer A.
      • Fasiolo F.
      • Hurt E.C.
      A conserved domain within Arc1p delivers tRNA to aminoacyl-tRNA synthetases.
      ,
      • Kyriacou S.V.
      • Deutscher M.P.
      An important role for the multienzyme aminoacyl-tRNA synthetase complex in mammalian translation and cell growth.
      ), subcellular location of aaRSs to control their noncanonical functions (
      • Golinelli-Cohen M.-P.
      • Mirande M.
      Arc1p is required for cytoplasmic confinement of synthetases and tRNA.
      ,
      • Arif A.
      • Jia J.
      • Mukhopadhyay R.
      • Willard B.
      • Kinter M.
      • Fox P.L.
      Two-site phosphorylation of EPRS coordinates multimodal regulation of noncanonical translational control activity.
      ,
      • Frechin M.
      • Enkler L.
      • Tetaud E.
      • Laporte D.
      • Senger B.
      • Blancard C.
      • et al.
      Expression of nuclear and mitochondrial genes encoding ATP synthase is synchronized by disassembly of a multisynthetase complex.
      ), and cellular turnover, by protecting associated proteins from degradation (
      • Han J.M.
      • Lee M.J.
      • Park S.G.
      • Lee S.H.
      • Razin E.
      • Choi E.-C.
      • et al.
      Hierarchical network between the components of the multi-tRNA synthetase complex.
      ).
      Figure thumbnail gr1
      Figure 1Structural diversity of eukaryotic MSCs. A, variability in proteins containing N-terminal GST and C-terminal EMAPII-like domains. S. cerevisiae Arc1p is monomeric (
      • Koehler C.
      • Round A.
      • Simader H.
      • Suck D.
      • Svergun D.
      Quaternary structure of the yeast Arc1p-aminoacyl-tRNA synthetase complex in solution and its compaction upon binding of tRNAs.
      ,
      • Deinert K.
      • Fasiolo F.
      • Hurt E.C.
      • Simos G.
      Arc1p organizes the yeast aminoacyl-tRNA synthetase complex and stabilizes its interaction with the cognate tRNAs.
      ), T. gondii Tg-p43 is dimeric (
      • van Rooyen J.M.
      • Murat J.-B.
      • Hammoudi P.-M.
      • Kieffer-Jaquinod S.
      • Coute Y.
      • Sharma A.
      • et al.
      Assembly of the novel five-component apicomplexan multi-aminoacyl-tRNA synthetase complex is driven by the hybrid scaffold protein tg-p43.
      ) and nothing is known about the oligomerization of T. brucei MCP1 (
      • Cestari I.
      • Kalidas S.
      • Monnerat S.
      • Anupama A.
      • Phillips M.A.
      • Stuart K.
      A multiple aminoacyl-tRNA synthetase complex that enhances tRNA-aminoacylation in african trypanosomes.
      ). The interactions between the two leucine-zippers of human AIMP1 and AIMP2 reconstitute a split protein with the same topology as Tg-p43 since it can homodimerize via the GST domain of AIMP2 (
      • Robinson J.-C.
      • Kerjan P.
      • Mirande M.
      Macromolecular assemblage of aminoacyl-tRNA synthetases: quantitative analysis of protein-protein interactions and mechanism of complex assembly.
      ). All of them are cytosolic while the plasmodial protein, tRip, is a dimer and is localized to the plasma membrane (
      • Bour T.
      • Mahmoudi N.
      • Kapps D.
      • Thiberge S.
      • Bargieri D.
      • Ménard R.
      • et al.
      Apicomplexa -specific tRip facilitates import of exogenous tRNAs into malaria parasites.
      ). B, MSC architectures. In S. cerevisiae, Arc1p binds to ERS and MRS and MSC assembly occurs only via GST-like domains (
      • Karanasios E.
      • Simader H.
      • Panayotou G.
      • Suck D.
      • Simos G.
      Molecular determinants of the yeast arc1p–aminoacyl-tRNA synthetase complex assembly.
      ). The T. gondii MSC is composed of Tg-p43 and four aaRSs: ERS, QRS, MRS, and YRS, where only YRS lacks a GST-like domain. T. brucei MSC contains three AIMPs (MCP1, MCP2, and MCP3) and at least six aaRSs (QRS, ARS, WRS, PRS, DRS, and MRS). Among these aaRSs, only MRS contains a GST-like domain. How the proteins associate in these two parasite MSCs is still unknown (indicated by red question mark). In the human MSC, AIMP2 is the component with the largest number of binding partners and is essential for complex assembly. Human MSC components are organized into two subcomplexes based on their association with AIMP2. Subcomplex I contains MRS, AIMP3, EPRS, IRS, LRS, KRS, and DRS and subcomplex II is composed of AIMP1, QRS, and RRS. aaRSs without GST-like domain are represented by pentagons, which are colored in dark gray when the enzyme is homodimeric. C, three types of interfaces can form between GST-like domains. Two canonical interaction modes are observed in GST-like dimers, which occur via interface 1 (through parallel helices α2 and α3) or interface 2 (through helices α7). P. vivax tRip dimerizes through an alternative interface 1 (
      • Gupta S.
      • Chhibber-Goel J.
      • Sharma M.
      • Parvez S.
      • Harlos K.
      • Sharma A.
      • et al.
      Crystal structures of the two domains that constitute the Plasmodium vivax p43 protein.
      ), where helices α2 and α3 of each monomer are oriented perpendicularly and contacts between α2–α2′ and α3–α3′ are observed. As a result, the β strands of the thioredoxin subdomain are oriented on the same side of the dimer. Interfaces 1 and 2 are indicated numerically throughout this figure. aaRS, aminoacyl-tRNA synthetase; AIMP, aaRS-interacting multifunctional protein; ARS, alanyl-tRNA synthetase; DRS, aspartyl-tRNA synthetase; ERS, glutamyl-tRNA synthetase; EPRS, glutamyl–prolyl-tRNA synthetase; IRS, isoleucyl-tRNA synthetase; KRS, lysyl-tRNA synthetase; LRS, leucyl-tRNA synthetase; MRS, methionyl-tRNA synthetase; MSC, multi-aaRS complex; PRS, prolyl-tRNA synthetase; RRS, arginyl-tRNA synthetase; QRS, glutaminyl-tRNA synthetase; tRip, tRNA import protein.
      The sizes and compositions of MSCs vary among organisms and usually reflect their biological complexities (
      • Laporte D.
      • Huot J.L.
      • Bader G.
      • Enkler L.
      • Senger B.
      • Becker H.D.
      Exploring the evolutionary diversity and assembly modes of multi-aminoacyl-tRNA synthetase complexes: Lessons from unicellular organisms.
      ) (Fig. 1B). However, despite their diversity, MSC assembly follows a common strategy involving GST-like modules found exclusively in eukaryotic aaRSs, AIMPs, and elongation factors EF1Bγ and EF1Bβ (
      • Havrylenko S.
      • Mirande M.
      Aminoacyl-tRNA synthetase complexes in evolution.
      ,
      • Koonin E.V.
      • Tatusov R.L.
      • Altschul S.F.
      • Bryant S.H.
      • Mushegian A.R.
      • Bork P.
      • et al.
      Eukaryotic translation elongation factor 1γ contains a glutathione transferase domain-Study of a diverse, ancient protein super family using motif search and structural modeling.
      ,
      • Bec G.
      • Kerjan P.
      • Waller J.P.
      Reconstitution in vitro of the valyl-tRNA synthetase-elongation factor (EF) 1 beta gamma delta complex. essential roles of the NH2-terminal extension of valyl-tRNA synthetase and of the EF-1 delta subunit in complex formation.
      ,
      • Guo M.
      • Yang X.L.
      • Schimmel P.
      New functions of aminoacyl-tRNA synthetases beyond translation.
      ,
      • Achilonu I.
      • Elebo N.
      • Hlabano B.
      • Owen G.R.
      • Papathanasopoulos M.
      • Dirr H.W.
      An update on the biophysical character of the human eukaryotic elongation factor 1 beta: Perspectives from interaction with elongation factor 1 gamma.
      ). For instance, the yeast MSC is a trimer where the GST-like domains of MRS, glutamyl-tRNA synthetase (ERS) and Arc1p associate (
      • Koehler C.
      • Round A.
      • Simader H.
      • Suck D.
      • Svergun D.
      Quaternary structure of the yeast Arc1p-aminoacyl-tRNA synthetase complex in solution and its compaction upon binding of tRNAs.
      ). In T. gondii, the MSC is composed of Tg-p43 and four aaRSs, MRS, ERS, glutaminyl-tRNA synthetase (QRS), and tyrosyl-tRNA synthetase (YRS) , where YRS is the only protein without a GST-like domain (
      • van Rooyen J.M.
      • Murat J.-B.
      • Hammoudi P.-M.
      • Kieffer-Jaquinod S.
      • Coute Y.
      • Sharma A.
      • et al.
      Assembly of the novel five-component apicomplexan multi-aminoacyl-tRNA synthetase complex is driven by the hybrid scaffold protein tg-p43.
      ). Among the six aaRSs (MRS, QRS, prolyl-tRNA synthetase (PRS), aspartyl-tRNA synthetase (DRS), tryptophanyl-tRNA synthetase (WRS), and alanyl-tRNA synthetase (ARS)) and the three AIMPs (MCP1, MCP2, and MCP3) that constitute the T. brucei MSC, MCP1, and MRS contain GST-like domains (
      • Cestari I.
      • Kalidas S.
      • Monnerat S.
      • Anupama A.
      • Phillips M.A.
      • Stuart K.
      A multiple aminoacyl-tRNA synthetase complex that enhances tRNA-aminoacylation in african trypanosomes.
      ). In the vertebrate MSC, the largest and most intricate MSC, crucial interactions involve GST-like domains and other types of interaction motifs such as leucine-zippers and WHEP domains (
      • Guo M.
      • Yang X.-L.
      Architecture and metamorphosis. In aminoacyl-tRNA synthetases in biology and medicine.
      ). Two AIMPs (AIMP2 and AIMP3 (p18)) and two aaRSs (MRS and bifunctional glutamyl–prolyl-tRNA synthetase (EPRS)) display GST-like domains. Together with AIMP1 and seven other aaRSs (QRS, DRS, lysyl-tRNA synthetase (KRS), arginyl-tRNA synthetase (RRS), isoleucyl-tRNA synthetase (IRS), and leucyl-tRNA synthetase (LRS)), they form a bisymmetrical complex (
      • Kaminska M.
      • Havrylenko S.
      • Decottignies P.
      • Gillet S.
      • Maréchal P.L.
      • Negrutskii B.
      • et al.
      Dissection of the structural organization of the aminoacyl-tRNA synthetase complex.
      ,
      • Mirande M.
      The aminoacyl-tRNA synthetase complex. In macromolecular protein complexes.
      ,
      • Khan K.
      • Baleanu-Gogonea C.
      • Willard B.
      • Gogonea V.
      • Fox P.L.
      3-Dimensional architecture of the human multi-tRNA synthetase complex.
      ).
      Several crystal structures of MSC subcomplexes involving GST-like domains are available. The GST-like fold consists of two subdomains (
      • Dirr H.
      • Reinemer P.
      • Huber R.
      X-ray crystal structures of cytosolic glutathione S-transferases. Implications for protein architecture, substrate recognition and catalytic function.
      ,
      • Armstrong R.N.
      Structure, catalytic mechanism, and evolution of the glutathione transferases.
      ) (Fig. 1C): the N-terminal thioredoxin-fold containing four β-strands (β1 to β4) and two helices (α1 and α2) and the C-terminal subdomain, which adopts an α-helical structure (α3 to α8). The central helix (α5) is mostly composed of hydrophobic residues and exhibits the N-capping box (S/T-X-X-D), which is strictly conserved and crucial for the stability of the fold (
      • Aceto A.
      • Dragani B.
      • Melino S.
      • Allocati N.
      • Masulli M.
      • Ilio C.D.
      • et al.
      Identification of an N-capping box that affects the α6-helix propensity in glutathione S-transferase superfamily proteins: a role for an invariant aspartic residue.
      ,
      • Cocco R.
      • Stenberg G.
      • Dragani B.
      • Principe D.R.
      • Paludi D.
      • Mannervik B.
      • et al.
      The folding and stability of human alpha class glutathione transferase A1-1 depend on distinct roles of a conserved N-capping box and hydrophobic staple motif.
      ). GST-like modules interact using two conserved interfaces. The dimerization interface 1 involves mainly polar residues within helices α2 and α3 of each GST-like monomer in parallel orientation, similar to what was observed in catalytically active GST enzymes (
      • Dirr H.
      • Reinemer P.
      • Huber R.
      X-ray crystal structures of cytosolic glutathione S-transferases. Implications for protein architecture, substrate recognition and catalytic function.
      ,
      • Armstrong R.N.
      Structure, catalytic mechanism, and evolution of the glutathione transferases.
      ). Dimerization through interface 2 involves the stacking of two arginines protruding from the α7 helix of each monomer and residues located in the loop connecting the α4 and α5 helices (
      • Simader H.
      • Hothorn M.
      • Köhler C.
      • Basquin J.
      • Simos G.
      • Suck D.
      Structural basis of yeast aminoacyl-tRNA synthetase complex formation revealed by crystal structures of two binary sub-complexes.
      ). Hence, Arc1p interacts with MRS through interface 1 (Protein Data Bank [PDB]: 2HSN) and with ERS through interface 2 (PDB: 2HRK) in the yeast MSC (
      • Simader H.
      • Hothorn M.
      • Köhler C.
      • Basquin J.
      • Simos G.
      • Suck D.
      Structural basis of yeast aminoacyl-tRNA synthetase complex formation revealed by crystal structures of two binary sub-complexes.
      ,
      • Karanasios E.
      • Simader H.
      • Panayotou G.
      • Suck D.
      • Simos G.
      Molecular determinants of the yeast arc1p–aminoacyl-tRNA synthetase complex assembly.
      ). The same network of interactions allows the formation of the GST-like heterotetramer, which builds the core of the human MSC (
      • Hahn H.
      • Park S.H.
      • Kim H.-J.
      • Kim S.
      • Han B.W.
      The DRS–AIMP2–EPRS subcomplex acts as a pivot in the multi-tRNA synthetase complex.
      ,
      • Cho H.Y.
      • Lee H.J.
      • Choi Y.S.
      • Kim D.K.
      • Jin K.S.
      • Kim S.
      • et al.
      Symmetric assembly of a decameric subcomplex in human multi-tRNA synthetase complex via interactions between glutathione transferase-homology domains and aspartyl-tRNA synthetase.
      ), with AIMP3 and MRS (PDB: 4BVX) or AIMP2 and EPRS (PDB: 5A34) dimerizing via interface 1 and AIMP3 and EPRS via interface 2 (PDB: 5BMU) (
      • Cho H.Y.
      • Maeng S.J.
      • Cho H.J.
      • Choi Y.S.
      • Chung J.M.
      • Lee S.
      • et al.
      Assembly of multi-tRNA synthetase complex via heterotetrameric glutathione transferase-homology domains.
      ).
      In Plasmodium, the protozoan parasite responsible for malaria, a protein named tRNA import protein (tRip) has been identified (
      • Bour T.
      • Mahmoudi N.
      • Kapps D.
      • Thiberge S.
      • Bargieri D.
      • Ménard R.
      • et al.
      Apicomplexa -specific tRip facilitates import of exogenous tRNAs into malaria parasites.
      ); it is a fusion protein between a GST-like domain and an EMAPII-like domain, like S. cerevisiae Arc1p, T. gondii Tg-p43, and T. brucei MCP1 (Fig. 1A). However, tRip is an integral membrane protein localized to the parasite plasma membrane. It is implicated in an unprecedented tRNA trafficking pathway and, when the tRip gene is deleted, the development of the blood stage parasite is significantly reduced (
      • Bour T.
      • Mahmoudi N.
      • Kapps D.
      • Thiberge S.
      • Bargieri D.
      • Ménard R.
      • et al.
      Apicomplexa -specific tRip facilitates import of exogenous tRNAs into malaria parasites.
      ). Recently, the crystal structures of both domains (N-terminal GST-like and C-terminal EMAPII-like) of Plasmodium vivax tRip were solved independently (
      • Gupta S.
      • Chhibber-Goel J.
      • Sharma M.
      • Parvez S.
      • Harlos K.
      • Sharma A.
      • et al.
      Crystal structures of the two domains that constitute the Plasmodium vivax p43 protein.
      ). The N-terminal GST-like domain of tRip forms two types of dimers (PDB: 5ZKF). Two monomers interact either through the canonical interface 2 or through an alternative interface 1. In this unusual conformation, α2 and α3 helices of one monomer pack against α2 and α3 helices of the second monomer, but they are oriented perpendicularly to each other (and not parallel as it is the case in a canonical interface 1) (Fig. 1C). This situation is unique among structurally characterized GST-like dimers.
      Thus, the combination between both its singular homodimeric conformation and its unusual role in tRNA import has raised questions about the implication of tRip in a Plasmodium MSC assembly. In this study, the existence of a tRip--bound MSC in the rodent malaria parasite Plasmodium berghei was experimentally verified. GST-like domains were used to determine their interaction patterns and infer the organization of the proteins in the parasite MSC. This approach led to an unexpected, novel observation: the interaction network between the different GST-like modules did not allow incorporation of all individual domains into a single MSC but rather led to the formation of two independent MSCs with specific biophysical properties in vitro.

      Results

      Identification of tRip partners in P. berghei blood stages

      tRip and its partners were extracted from P. berghei blood-stage parasites using a purified and specific antibody raised against the C-terminal EMAPII-like domain of Plasmodium falciparum tRip (
      • Bour T.
      • Mahmoudi N.
      • Kapps D.
      • Thiberge S.
      • Bargieri D.
      • Ménard R.
      • et al.
      Apicomplexa -specific tRip facilitates import of exogenous tRNAs into malaria parasites.
      ). The strong lysis conditions resulting from saponin treatment significantly reduced the number of spectra in the b samples leading to its exclusion from the bioinformatics analysis. An average of 250 P. berghei proteins were thus identified in samples a, c, and d (three WT and three KO). As expected, tRip was found only in the WT samples. After subtraction of background interactions and protein frequency assessment, only four proteins were retained (ERS, MRS, QRS, and a nuclear ribosomal biogenesis regulatory protein) and considered statistically significantly increased in the WT sample, with an adjusted p value < 0.05, a minimum relevant spectral count of 5 (average), and a min Log fold change of 2 (Fig. 2A and Table S1). One candidate, ERS, stood out since its deltaSC and p value parameters were identical to those of tRip. Because of its nuclear localization and lack of detection in sample d, the ribosomal biogenesis regulatory protein was not considered further. Only one protein was significantly increased in the KO parasite (60S ribosomal protein L44, putative) compared to the WT.
      Figure thumbnail gr2
      Figure 2Identification of the tRip interactome in the P. berghei blood stage. A, volcano plot visualization of the tRip interactome. CoIP and mass spectrometry analysis (LC-MS/MS) of tRip partners were performed in three biological replicates, and LC-MS/MS data were analyzed by spectral counts. All identified proteins that were detected with more than five spectra were used for the generation of the differential expression test. Three aaRSs (2, 3, and 4) specifically bound to tRip (1) in all three samples with p values < 0.05, all of which have N-terminal GST-like domains. A nuclear protein involved in the ribosome biogenesis (5) was observed in two samples (a) and (c), and a ribosomal protein (6) was observed in the tRip-KO parasite in samples (a) and (d). Complete results are presented in . B, schematic representation of tRip and partners. aaRS cores are shown in black (ERS), blue (QRS), and orange (MRS), GST-like domains are shown in green, RNA-binding domains are shown in gray, and LCRs are shown in pale yellow; K indicates the K-rich C-terminal domain of Plasmodium QRS. aaRS, aminoacyl-tRNA synthetase; coIP, coimmunoprecipitation; ERS, glutamyl-tRNA synthetase; MRS, methionyl-tRNA synthetase; QRS, glutaminyl-tRNA synthetase; tRip, tRNA import protein.

      Structural organization of selected aaRSs

      The sequences of P. berghei tRip, ERS, QRS, and MRS were analyzed using Blast and multisequence alignments (MSAs) (Fig. 2B). ERS, QRS, and MRS are all class I aaRSs characterized by HIGH and KMSKS motifs in their catalytic domains. Compared to their prokaryotic homologs, they contain additional sequences appended to their N- and C-terminal ends and several Plasmodium-specific insertions referred as low-complexity regions (LCRs). The length and composition of LCRs are generally variable between Plasmodium species and strains, but their localization is conserved within homologous proteins (
      • Frugier M.
      • Bour T.
      • Ayach M.
      • Santos M.A.
      • Rudinger-Thirion J.
      • Theobald-Dietrich A.
      • et al.
      Low Complexity Regions behave as tRNA sponges to help co-translational folding of plasmodial proteins.
      ,
      • Chaudhry S.R.
      • Lwin N.
      • Phelan D.
      • Escalante A.A.
      • Battistuzzi F.U.
      Comparative analysis of low complexity regions in Plasmodia.
      ). The longest LCRs are mainly observed in P. falciparum proteins and are characterized by long asparagine (N) repeats (
      • Singh G.P.
      • Chandra B.R.
      • Bhattacharya A.
      • Akhouri R.R.
      • Singh S.K.
      • Sharma A.
      Hyper-expansion of asparagines correlates with an abundance of proteins with prion-like domains in Plasmodium falciparum.
      ).
      P. berghei tRip, ERS, QRS, and MRS sequences were then submitted to the Raptor X web server (http://raptorx.uchicago.edu/) (
      • Källberg M.
      • Wang H.
      • Wang S.
      • Peng J.
      • Wang Z.
      • Lu H.
      • et al.
      Template-based protein structure modeling using the RaptorX web server.
      ) for 3D structure prediction. The Raptor X models were consistent with the Blast and the MSA analysis, and the top templates used to model each domain are shown in Table S2. All three aaRSs contained an N-terminal GST-like domain; MRS contained an EMAPII-like domain appended to its C terminus, while QRS had a C-terminal extension that was too short to be considered as an independent domain by Raptor X. Nevertheless, this extension contains many positively charged residues and has the potential to form one or more helices, suggesting RNA-binding properties (
      • Frugier M.
      • Moulinier L.
      • Giegé R.
      A domain in the N-terminal extension of class IIb eukaryotic aminoacyl-tRNA synthetases is important for tRNA binding.
      ,
      • Francin M.
      • Mirande M.
      Functional dissection of the eukaryotic-specific tRNA-interacting factor of lysyl-tRNA synthetase.
      ,
      • Crepin T.
      • Peterson F.
      • Haertlein M.
      • Jensen D.
      • Wang C.
      • Cusack S.
      • et al.
      A hybrid structural model of the complete Brugia malayi cytoplasmic asparaginyl-tRNA synthetase.
      ) (Fig. S1, AF). The prediction of C-terminal domains with potential tRNA-binding abilities in MRS and QRS led us to test whether these domains could bind tRNAs in vitro. Indeed, the presence of these C-terminal domains increased the interaction of the MRS and QRS anticodon-binding domains toward total human tRNA (Fig. S1, G and H), indicating that their binding were sequence nonspecific. These extensions are also conserved in MRS and QRS from other Apicomplexan parasites such as T. gondii (
      • van Rooyen J.M.
      • Murat J.-B.
      • Hammoudi P.-M.
      • Kieffer-Jaquinod S.
      • Coute Y.
      • Sharma A.
      • et al.
      Assembly of the novel five-component apicomplexan multi-aminoacyl-tRNA synthetase complex is driven by the hybrid scaffold protein tg-p43.
      ).

      GST-like domains mediate Plasmodium MSC assembly

      Production of full-length ERS, QRS, and MRS led to low expression, limited solubility, or proteolysis of the recombinant enzymes, and mass spectrometry (MS) analyses revealed that recombinant ERS and QRS lacked their N-terminal GST-like domains. Moreover, none of these purified recombinant proteins interacted with tRip (Fig. S2). We therefore chose to focus specifically on the GST-like domains from each aaRS to investigate their interactions with each other and tRip. From here on, these domains are referred to as ERS-N (ERS1-228), QRS-N (QRS1-208), and MRS-N (MRS1-228). Several constructs with and without a C-terminal 6His (-6H) or SUMO-6His (S-6H) tag were designed (Table S3). The SUMO domain not only slightly improved the solubility of MRS-N and QRS-N, but it also provided proteins of different sizes that could be distinguished one from each other on SDS-PAGE. Each partner was alternatively used as bait or prey in pull-down assays. In these assays, mixture of bacteria expressing (i) the 6His-tagged bait protein and (ii) one or more prey proteins was lysed and the protein extract was incubated with nickel-nitrilotriacetic acid (Ni-NTA) resin. This resin binds the bait protein and indirectly captures any prey protein through noncovalent interactions with the bait. After incubation, the resin was washed thoroughly, and the bound proteins were eluted and analyzed by SDS-PAGE. If an interaction occurred, bait and prey proteins appear together in the elution fraction.
      Pairwise interactions showed that ERS-N associates with the other three domains (Figs. 3A and S3A); that MRS-N interacts with QRS-N and that MRS-N and ERS-N homo-oligomerize. As expected, the C-terminal domain of tRip was dispensable for its interaction with ERS-N, indicating that its GST-like domain is sufficient for complex assembly. However, the presence of the 6His tag directly attached at the ERS-N C terminus reduces its ability to interact with QRS-N. Similarly, a 6His tag directly linked to the C terminus of MRS-N abolishes any pairwise interaction. Thus, subsequent pull-down assays were performed with SUMO-tagged baits.
      Figure thumbnail gr3
      Figure 3Interaction between P. berghei tRip, ERS, MRS, and QRS GST-like domains. A, summary of pairwise interactions. Results of pull-down experiments performed with two GST-like domains (A) are shown (n ≥ 3). The thick arrows indicate the most effective interactions and the thin arrows the weakest interactions. B, pull-down experiments with four proteins. Each protein was used as bait to test for capture of the other three (n ≥ 3). The content of the initial mixture and the captured proteins are shown on each gel. Partners are identified in gray (tRip), black (ERS-N-S or ERS-N), blue (QRS-N-S or QRS-N), and orange (MRS-N-S or MRS-N) arrows; the bait protein is indicated by a red asterisk. MRS-N-S and tRip as well as ERS-N and QRS-N comigrate. The additional bands present on the gels could correspond to either E. coli EF-Tu or outer membrane proteins A; these proteins are about 40 kDa and are among the most abundant contaminants (). C, competitions; series of pull-down experiments were performed with constant concentrations of QRS-N-S-6H–tRip–ERS-N or MRS-N-S-6H–tRip:ERS-N and increasing concentrations of MRS-N or QRS-N, respectively. D, specificity controls; the GST-like domain present at the N terminus of human glutamyl-prolyl-tRNA synthetase (EPRS) was fused to the SUMO-6H tag (HsEPRS-N-S-6H) (empty black arrow) and replaced the ERS-N-S-6H bait in the pull-down experiment with four partners. Alternatively, the N-terminal GST-like domain of P. berghei EF1Bγ (empty green arrow) was used as prey in place of MRS-N, QRS-N, and tRip. E, segregation of complexes upon tRNA binding. The four proteins were copurified using ERS-N-S as bait and tested for their capacity to bind yeast total tRNA. With a tRNA:complex ratio of 1:1, assuming the complex is monomeric, three bands (∗, ∗∗, and ∗∗∗) were observed in a 1% agarose gel after RNA and protein visualization with ethidium bromide and Coomassie blue staining, respectively. The bands were cut out and their protein content was analyzed by SDS-PAGE (n = 3). ERS, glutamyl-tRNA synthetase; MRS, methionyl-tRNA synthetase; QRS, glutaminyl-tRNA synthetase; tRip, tRNA import protein.
      Pull-down assays with all four partners were performed, using alternative baits (6H-tRip, ERS-N-S-6H, QRS-N-S-6H, and MRS-N-S-6H). Individually, 6H-tRip and ERS-N-S-6H allowed the pull down of the other three proteins showing that each of these proteins interacts with all MSC components (Fig. S3B). In contrast, QRS-N-S-6H and MRS-N-S-6H led to the efficient capture of tRip and ERS-N but also to the apparent loss of the fourth partner (Fig. 3B). MS analysis corroborated the profile observed on the analytical gels. While MRS-N and QRS-N are present in comparable amounts with ERS-N-S-6 as the bait, this equilibrium is disrupted when the bait is either MRS-N-S-6H or QRS-N-S-6H (bar charts in Fig. S3B and data in Table S1). Because it is difficult to control the amount of each recombinant protein in the samples, we performed competition experiments between QRS-N and MRS-N for binding to the tRip–ERS-N subcomplex. In these assays, tRip–ERS-N–MRS-N-S-6H and tRip–ERS-N–QRS-N-S-6H ternary complex were challenged with increasing concentrations of competitors, QRS-N or MRS-N, respectively (Fig. 3C). In both cases, we observe not only that the competitor does not integrate a quaternary complex but also that tRip–ERS-N gradually vanishes in the presence of the highest competitor concentrations (Fig. S3C), suggesting the formation of the alternative ternary complex lacking a 6His tag and thus eliminated during washes. Additional controls were performed with two other GST-like domains (Fig. 3D). On the one hand, the N-terminal GST-like domain of human EPRS fused to the SUMO-6H tag (HsEPRS-N-S-6H) was used in place of ERS-N-S-6H as a bait. On the other hand, the N-terminal GST-like domain of P. berghei EF1Bγ was tested as a prey for its ability to substitute for MRS-N, QRS-N, or tRip in pull-down experiments. None of these control domains could be incorporated into the complexes, indicating that the SUMO module does not promote the integration of nonspecific GST-like domains and that the interactions that drive complex formation are highly specific.
      Finally, triplicates of large-scale pull-down experiments were performed using ERS-N-S-6H, QRS-N-S-6H, or MRS-N-S-6H as baits to further purify the different complexes by size-exclusion chromatography (SEC) and analyze them by MS (Fig. 4). In all cases, the captured proteins eluted as a main peak with comparable elution volumes on SEC, indicating the formation of soluble complexes with apparent molecular weights (MWs) of 300 to 400 kDa (Fig. 4A). Pull down with either QRS-N-S-6H or MRS-N-S-6H led to the purification of heterotrimeric complexes that excluded MRS-N or QRS-N, respectively (Fig. 4, B and C). Although a heterotetrameric complex cannot be ruled out, these observations are consistent and strongly suggest the existence of two mutually exclusive heterotrimeric complexes in vitro: tRip–ERS-N–QRS-N and tRip–ERS-N–MRS-N. For convenience, these heterotrimers are referred as Q-complex and M-complex, respectively.
      Figure thumbnail gr4
      Figure 4Composition of complexes. Bacteria expressing the tRip and other GST-like partners were lysed together and protein extracts were subjected to Ni-affinity and size-exclusion chromatographies (SECs). Different baits (ERS-N-S-6H, QRS-N-S-6H, or MRS-N-S-6H) were used to capture the different preys (tRip, ERS-N, QRS-N, and MRS-N). A, SEC chromatograms. Complexes eluted as a main peak on a SepFast 6 to 5000 kDa (n > 3) column. Graphs show the elution profile (left axis) and calibration curve (right axis, R2 = 0.98) based on the elution of the following size markers: thyroglobulin (669 kDa), γ-globulin (158 kDa), ovalbumin (44 kDa), myoglobin (17 kDa), and vitamin B12 (1.35 kDa). B, SDS-PAGE analysis. Fractions of the main SEC peak were analyzed by SDS-PAGE. C, mass spectrometry. Fractions marked with arrows (in panel B) were pooled and analyzed by mass spectrometry. Three independent experiments were performed, and the relative abundance of each partner (based on the number of spectra) was plotted. Colors were maintained: gray for tRip, black for ERS-N, blue for QRS-N, and orange for MRS-N. Error bars represent SD across the triplicate measurements. Purification of complexes on SEC led to more homogeneous samples for mass spectrometry analysis. ERS, glutamyl-tRNA synthetase; MRS, methionyl-tRNA synthetase; QRS, glutaminyl-tRNA synthetase; tRip, tRNA import protein.

      Oligomeric state of Q- and M-complexes in solution

      Q- and M-complexes were purified with QRS-N-S-6H and MRS-N-S-6H baits, respectively. SUMO fusions were proteolytically removed before the final purification step on a calibrated SEC column (Fig. 5A). As expected, the main peaks contained heterotrimeric complexes consisting of equivalent amounts of each partner, as assessed by SDS-PAGE (Fig. 5B) and further confirmed that the SUMO-tag is not involved in the complex association. The apparent MWs of the Q- and M-complexes derived from the SEC profiles (about 250 and 275 kDa, respectively), indicated that both complexes (with theoretical MWs of 97 kDa and 101.5 kDa, respectively) were elongated or oligomerized. Thus, their size and homogeneity were further investigated by batch light-scattering measurements. On the one hand, dynamic light scattering (DLS) indicated homogeneous samples (Fig. 5C) and measured MW-R (MW based on hydrodynamic radius) of 295 and 393 kDa assuming spherical particles for the Q- and M-complexes, respectively. On the other hand, static light scattering (SLS) yielded MW-S (absolute MW) of 204 and 247 kDa for the two complexes, respectively. These measurements indicate not only that the complexes contained at least two copies of each partner but also that they had an elongated conformation (MW-R/MW-S > 1). Only SEC–multiangle light scattering (MALS) could establish more precisely the oligomeric state of the complexes (Fig. 5D). While the Q-complex yielded a MW of 187.3 kDa, close to the expected value for a dimeric heterotrimer (193.6 kDa), the M-complex showed overlapping oligomeric species. The major peak indicated a value of about 217 kDa, corresponding to the theoretical MW of a dimeric heterotrimeric M-complex (203 kDa), yet an additional population was not completely separated by the column and was characterized by a larger size around 500 kDa. This size estimation corresponds to two to three copies of the dimeric ternary complex, suggesting that the M-complex might oligomerize further in solution. This observation was supported by light-scattering experiments on MRS-N, which showed that at a concentration suitable for measurements (i.e., 1.8 mg/ml), most MRS-N-S-6H molecules appear as elongated dimers; yet, populations with higher apparent sizes were visualized by SEC and DLS/SLS analysis (Fig. S4A).
      Figure thumbnail gr5
      Figure 5Purification and characterization of Plasmodium Q- and M-complexes. Q- and M-ternary complexes were purified using QRS-N-S-6H and MRS-N-S-6H as respective baits in the presence of tRip and ERS-N. A, SEC experiments. The SUMO tags were removed by TEV cleavage, and the complexes were applied to a calibrated SepFast 6 to 5000 kDa column (n > 3). Graphs show the elution profile (left axis) and calibration curve (right axis, R2 >0.9) based on the elution of the following size markers: thyroglobulin (669 kDa), apoferritin (443 kDa), β-amylase (200 kDa), alcohol dehydrogenase (150 kDa), bovine serum albumin (66 kDa), and carbonic anhydrase (29 kDa). B, fraction analysis by SDS-PAGE and quantification. The relative amount of each partner was quantified in the SEC peaks from (A). Fractions containing equimolar amounts of domains were pooled and used in light-scattering measurements. C, DLS/SLS experiments. Graphs show the intensity-based particle size distribution (left y-axis) as a function of particle radius (lower x-axis), and the experimental data (black dots) as well as the cumulant-fitted autocorrelation function (red line) (right y-axis) as a function of time (upper x-axis). The Q- and M-complexes were assayed at concentrations of 3.7 and 4.1 mg/ml, respectively. Hydrodynamic radius (Rh), polydispersity (PD), molecular weight (MW-R) derived from Cumulants DLS analysis, and the absolute molecular weight (MW-S) inferred by SLS are indicated. D, SEC-MALS experiments. Elution profiles of Q- and M-complexes on a Superose 6 column are shown. Normalized light scattering (LS, orange) and differential refractive index (dRI, blue) are on the left y-axis and molecular weights (gray) on the right y-axis. While the Q complex is homogeneous and appears as a dimer, SEC separation did not provide dimer–oligomer resolution for the M-complex (n = 2). DLS, dynamic light scattering; ERS, glutamyl-tRNA synthetase; MALS, multiangle light scattering; MRS, methionyl-tRNA synthetase; QRS, glutaminyl-tRNA synthetase; SLS, static light scattering; tRip, tRNA import protein.

      Two types of ternary GST-like complexes with different tRNA-binding profiles

      EMSAs with individual Q- and M-complexes (purified with QRS-N-S-6H or MRS-N-S-6H) in the presence of yeast total tRNA revealed specific migration profiles (Fig. S5). Control experiments using complexes lacking the EMAPII-like domain (tRip-N) confirmed that tRNA binding was specific to the C-terminal domain of tRip. Furthermore, all three proteins were present in the tRNA-bound fraction (Fig. S5), indicating that the complexes did not dissociate upon tRNA binding.
      Alternatively, in EMSA performed with the complexes purified with ERS-N-S-6H bait in the presence of tRip, QRS-N and MRS-N also shifted the tRNAs but showed three populations (Fig. 3E). The corresponding bands were cut out and analyzed by SDS-PAGE to determine their protein content. The upper band (∗) contained only the Q-complex while the bottom band (∗∗∗) contained only the M-complex. This segregation upon tRNA binding supported our hypothesis that the proteins copurified with ERS-N-S-6H form two independent heterotrimeric complexes. Yet, all four proteins were observed in the middle band (∗∗), probably because of an overlap of the two ternary complexes.

      Mapping interaction interfaces in Q- and M-complexes

      In the presence of two exclusive heterotrimeric complexes organized around the tRip–ERS-N heterodimer and containing either QRS-N or MRS-N, we chose to test the effect of point mutations in pull-down experiments with ERS-N-S-6H as the bait and all three preys (tRip, QRS-N, and MRS-N) together. This approach allowed us to assess whether the mutations had a global impact on both complexes or had a specific effect on the formation of a given one (Q or M). The list of mutations and their effect on pull-down assays are summarized in Figure 6A (list of mutations), Figure 6B (results summary), and Figs. S6 (sequence alignments) and S7 (experimental data).
      Figure thumbnail gr6
      Figure 6Arrangement of GST-like domains in Q- and M-ternary complexes. A, probing of GST–GST interactions between tRip, ERS-N, QRS-N, and MRS-N. Bold residues correspond to conserved residues present in helices 2, 3, and 7 of the N-terminal GST-like domains of tRip, ERS, MRS, and QRS. The effects of the mutations are indicated with different symbols: () no effect on heterodimerization; () the replacement of a small uncharged amino acid (alanine, leucine, or valine) with a large charged amino acid (arginine) interfered with the proximity between GST domains; () alanine replacements disrupted the interaction between GST domains. Gray boxes indicate that amino acid mutation led to the complete destabilization of the mutated protein and (D) indicates amino acids that are involved in homodimerization. Each interaction was tested by pull-down assays at least three times. B, modeling of the GST-like interfaces. All amino acids tested are shown on the crystal structure of P. vivax tRip (PDB: 5ZKE) and Raptor X models of ERS-N, QRS-N, and MRS-N. Models are shown in light gray and the domains investigated by mutagenesis are shown in dark gray (tRip), black (ERS-N), blue (QRS-N), or orange (MRS-N). Silent mutations are shown in green and the mutations leading to the disruption of the complexes are labeled and shown with red sticks C, schematic representation of the two ternary complexes identified in P. berghei. Models based on this study show two complexes containing two copies each of membrane-bound tRip, ERS-N, and either QRS-N (Q-complex) or MRS-N (M-complex). The interaction interfaces of each partner according to mutagenesis experiments are indicated. Inclusion of MRS-N in the M-complex depended on both tRip dimerization and tRip–ERS-N interaction (dashes). ERS, glutamyl-tRNA synthetase; MRS, methionyl-tRNA synthetase; PDB, Protein Data Bank; QRS, glutaminyl-tRNA synthetase; tRip, tRNA import protein.
      tRip is a homodimer in solution, as shown by light-scattering analysis (MW-S = 114 kDa, only slightly larger than the calculated 92.4 kDa). In addition, the shape factor deduced from these measures (MW-R/MW-S > 1) suggests that tRip is an elongated dimer (Fig. S4B), which associates via a unique interaction as seen in the crystal structure of tRip-N (
      • Gupta S.
      • Chhibber-Goel J.
      • Sharma M.
      • Parvez S.
      • Harlos K.
      • Sharma A.
      • et al.
      Crystal structures of the two domains that constitute the Plasmodium vivax p43 protein.
      ) (Fig. 1C). This alternative interface 1 mainly involves phenylalanine residues: F58 in helix α2 and F90 in helix α3 (Fig. S6A). Independent mutation F58A and F90A did not impair tRip–ERS-N heterodimerization or the formation of the Q-complex but specifically hindered the association of MRS-N into the M-complex.
      Among the four GST-like partners, only tRip and ERS-N contain a strictly conserved arginine in their respective α7 helices (Fig. S6B), suggesting that they might interact via a canonical interface 2 (Fig. S6A). As expected, tRip R154A and ERS-N-S-6H R198A completely abolished interactions between the two partners. Yet, the R198A mutation in ERS-N-S-6H also specifically inhibited the binding of MRS-N in the complex, indicating that ERS-N needs to bind tRip to bring MRS-N into the M-complex. In contrast, the dissociation between tRip and ERS-N did not affect binding between QRS-N and ERS-N-S-6H in the Q-complex.
      To initiate the investigation of interfaces connecting ERS-N and QRS-N or MRS-N, we used the same approach as (
      • Simader H.
      • Hothorn M.
      • Köhler C.
      • Basquin J.
      • Simos G.
      • Suck D.
      Structural basis of yeast aminoacyl-tRNA synthetase complex formation revealed by crystal structures of two binary sub-complexes.
      ,
      • Cho H.Y.
      • Lee H.J.
      • Choi Y.S.
      • Kim D.K.
      • Jin K.S.
      • Kim S.
      • et al.
      Symmetric assembly of a decameric subcomplex in human multi-tRNA synthetase complex via interactions between glutathione transferase-homology domains and aspartyl-tRNA synthetase.
      ,
      • Cho H.Y.
      • Maeng S.J.
      • Cho H.J.
      • Choi Y.S.
      • Chung J.M.
      • Lee S.
      • et al.
      Assembly of multi-tRNA synthetase complex via heterotetrameric glutathione transferase-homology domains.
      ). Small uncharged amino acids (alanine (A), valine (V), isoleucine (I), or leucine (L)), when present on the surface of helices α2 and α3, were replaced with the large and charged arginine (R) to disrupt any proximity contacts between potentially interacting surfaces. In ERS-N, only A124R (α3) interfered with both Q- and M-complexes, whereas I89R (α2) specifically disrupted the Q-complex, and L97R (α2) and V120R (α3) specifically disrupted the M-complex, suggesting that QRS-N and MRS-N recognized ERS-N interface 1 but do not share interaction patterns. Since the GST-like canonical interface 1 is globally polar, as shown by the crystal structure of the human MRS–AIMP3 and EPRS–AIMP2 heterodimers and, to a lesser degree, that of the yeast MRS–Arc1p heterodimer (Fig. S6A), we looked for conserved polar residues in helices α2 and α3 in ERS-N. Mutations of D87A or D95A in α2 destabilized ERS-N, while R134A (α3) did not have any effect on Q- or M-complex formation. E128A and D131A mutations in α3 disrupted only the Q-complex.
      In the mirror experiment R replacements of V66 (α2) or V99 (α3) as well as mutations of N62A (α2), K105A (α3), and K109A (α3) in QRS-N, disrupted QRS-N uptake into the Q-complex (Fig. 6A). The participation of so many polar residues in the ERS-N–QRS-N interaction is reminiscent of what has been shown in human heterodimers, especially between EPRS and AIMP2 (
      • Cho H.Y.
      • Maeng S.J.
      • Cho H.J.
      • Choi Y.S.
      • Chung J.M.
      • Lee S.
      • et al.
      Assembly of multi-tRNA synthetase complex via heterotetrameric glutathione transferase-homology domains.
      ).
      Association between ERS-N and MRS-N was less obvious. Indeed, only an A65R mutation in MRS-N α2 disrupted its interaction with the M-complex, and the mutation of most of the polar amino acids present in α3 (D98A, E101A, and K102A) had no effect, except for E95A. Helices α2 and α3 contain several aromatic residues (Fig. S6B), as it is the case in tRip α2 and α3 (
      • Gupta S.
      • Chhibber-Goel J.
      • Sharma M.
      • Parvez S.
      • Harlos K.
      • Sharma A.
      • et al.
      Crystal structures of the two domains that constitute the Plasmodium vivax p43 protein.
      ). Only mutations of F68A (α2), H71A (α2), or Y90A (α3) in MRS-N and of F90A (α2) in ERS-N-S-6H led to the specific disruption of the M-complex, suggesting that MRS-N might interact with ERS-N via an alternative interface 1 (Figs. 6 and S6A).

      Discussion

      In this study, we immunoprecipitated, under stringent selection conditions, proteins interacting with the endogenous P. berghei tRip. The identified interactome only contained aaRSs. These aaRSs catalyzed the specific aminoacylation of tRNAs with glutamate (ERS), methionine (MRS), and glutamine (QRS) (Fig. 2A). As in tRip, all three Plasmodium aaRSs contain GST-like domains at their N termini (Fig. 2B). Only six Plasmodium proteins containing a GST-like domain were identified in the conserved domain database (CDD) (
      • Marchler-Bauer A.
      • Bo Y.
      • Han L.
      • He J.
      • Lanczycki C.J.
      • Lu S.
      • et al.
      CDD/SPARCLE: Functional classification of proteins via subfamily domain architectures.
      ). They corresponded to tRip, ERS, MRS, the elongation factor 1β (EF1β, PBANKA_081420), the GST-1 (PBANKA_10239), and the thioredoxin-GST-EF1Bγ (PBANKA_135200). The N-terminal GST-like domain of QRS was not considered as such in the CDD, probably due to the low sequence conservation and the presence of an LCR in the middle of the structural fold. However, further investigations, including its modeling with Raptor X, allowed its detection (Fig. 2 and Table S2).
      Based on aaRSs composition, the closest MSCs to the Plasmodium MSC were those from S. cerevisiae (
      • Galani K.
      • Großhans H.
      • Deinert K.
      • Hurt E.C.
      • Simos G.
      The intracellular location of two aminoacyl-tRNA synthetases depends on complex formation with Arc1p.
      ) or T. gondii (
      • van Rooyen J.M.
      • Murat J.-B.
      • Hammoudi P.-M.
      • Kieffer-Jaquinod S.
      • Coute Y.
      • Sharma A.
      • et al.
      Assembly of the novel five-component apicomplexan multi-aminoacyl-tRNA synthetase complex is driven by the hybrid scaffold protein tg-p43.
      ), both of which have been characterized as cytosolic. The purification of T. gondii MSC led to significantly heterogeneous samples in size and composition (
      • van Rooyen J.M.
      • Murat J.-B.
      • Hammoudi P.-M.
      • Kieffer-Jaquinod S.
      • Coute Y.
      • Sharma A.
      • et al.
      Assembly of the novel five-component apicomplexan multi-aminoacyl-tRNA synthetase complex is driven by the hybrid scaffold protein tg-p43.
      ). In contrast, the S. cerevisiae MSC is well defined and interactions between the three GST-like domains have been identified by crystallography (
      • Koehler C.
      • Round A.
      • Simader H.
      • Suck D.
      • Svergun D.
      Quaternary structure of the yeast Arc1p-aminoacyl-tRNA synthetase complex in solution and its compaction upon binding of tRNAs.
      ,
      • Simader H.
      • Hothorn M.
      • Köhler C.
      • Basquin J.
      • Simos G.
      • Suck D.
      Structural basis of yeast aminoacyl-tRNA synthetase complex formation revealed by crystal structures of two binary sub-complexes.
      ). The present study indicates that P. berghei harbors two MSCs, the Q- and M-complexe, each containing two copies of tRip, ERS-N and QRS-N or of tRip, ERS-N, and MRS-N, respectively. ERS-N has a central position, since it binds simultaneously to tRip, QRS-N, or MRS-N. Nevertheless, QRS-N and MRS-N diverge in their interaction pattern with ERS-N. While the association of MRS-N in the M-complex requires the presence of a dimeric tRip associated to ERS-N, QRS binds directly to ERS-N even in the absence of tRip (Fig. 6A). Moreover, mutational experiments confirmed that QRS-N and MRS-N bind to the same interface on ERS-N, but with two different strategies (Fig. 6A). Our favorite hypothesis is that interactions between QRS-N and ERS-N would mainly involve polar residues as in a GST–GST canonical interface 1 and that binding of MRS-N to ERS-N–tRip would use a different pattern closer to an alternative interface 1, with an aromatic environment (Fig. 6).
      In yeast, the isolated EMAPII-like domain of Arc1p binds with high affinity and specificity to tRNAMet and tRNAGlu and strongly increases the affinity of MSC-associated MRS and ERS for their cognate tRNAs (
      • Simos G.
      • Sauer A.
      • Fasiolo F.
      • Hurt E.C.
      A conserved domain within Arc1p delivers tRNA to aminoacyl-tRNA synthetases.
      ,
      • Deinert K.
      • Fasiolo F.
      • Hurt E.C.
      • Simos G.
      Arc1p organizes the yeast aminoacyl-tRNA synthetase complex and stabilizes its interaction with the cognate tRNAs.
      ). In contrast, the EMAPII-like domain of tRip is located on the parasite surface and thus theoretically cannot participate in tRNA aminoacylation within the parasite. The absence of this domain in the vicinity of the aaRSs could be balanced by the presence of the C-terminal extensions of Plasmodium QRS and MRS (Figs. 2 and S1). Indeed, these extensions had nonspecific RNA-binding capacities, although characterized by low affinities (Fig. S1H). The tRNA-binding capacity is generally low for such isolated domains, whether it is the human AIMP1 domain (
      • Quevillon S.
      • Agou F.
      • Robinson J.-C.
      • Mirande M.
      The p43 component of the mammalian multi-synthetase complex is likely to Be the precursor of the endothelial monocyte-activating polypeptide II cytokine.
      ), the Oryza sativa C-terminal MRS domain (
      • Kaminska M.
      • Deniziak M.
      • Kerjan P.
      • Barciszewski J.
      • Mirande M.
      A recurrent general RNA binding domain appended to plant methionyl-tRNA synthetase acts as a cis-acting cofactor for aminoacylation.
      ), or the WHEP domain of the human EPRS (
      • Cahuzac B.
      A recurrent RNA-binding domain is appended to eukaryotic aminoacyl-tRNA synthetases.
      ). Yet, when fused to the core of aaRSs, these domains significantly increase the affinity of the enzymes for their cognate tRNAs. By analogy, the C-terminal domains of Plasmodium MRS and QRS could act synergistically and confer the ability to bind their cognate tRNAs with higher affinities to these enzymes to compensate for the membrane localization of tRip.
      The EMAPII-like domain of P. falciparum tRip exhibits a unique specificity pattern. It discriminates tRNAs based on their posttranscriptional modification profiles, and tRip's preferred binders are isoacceptors of human host tRNAAla, tRNASer, tRNALeu, and tRNAAsn (
      • Cela M.
      • Théobald-Dietrich A.
      • Rudinger-Thirion J.
      • Wolff P.
      • Geslain R.
      • Frugier M.
      Identification of host tRNAs preferentially recognized by the Plasmodium surface protein tRip.
      ), suggesting that import of exogenous tRNAs is not related to aminoacylation by MSCs’ aaRSs. It is worth noticing that the single human tRNAAsn isoacceptor is not only one of the best ligands for tRip but also that asparagine is the predominant amino acid in the P. falciparum proteome. Most asparagine residues are found in LCRs and form long homopolymeric regions (
      • Chaudhry S.R.
      • Lwin N.
      • Phelan D.
      • Escalante A.A.
      • Battistuzzi F.U.
      Comparative analysis of low complexity regions in Plasmodia.
      ) suggesting that tRNA import may facilitate their synthesis. Although asparagine-rich insertions are present in virtually all Plasmodium proteins, their role is not yet established; these repeats are suspected to be involved in protein–protein interactions and/or aggregation, protein localization, immune response, or parasite virulence (
      • Davies H.M.
      • Nofal S.D.
      • McLaughlin E.J.
      • Osborne A.R.
      Repetitive sequences in malaria parasite proteins.
      ).
      The two Plasmodium MSCs could differ in their ability to organize in the parasite membrane and might thus behave differently in tRNA aminoacylation, tRNA import, or any other alternative function of the associated aaRSs. Indeed, biophysical experiments showed that the M-complex is characterized by larger apparent sizes than those determined for the Q-complex (Fig. 5). Interestingly, pairwise pull-down assays and DLS/SLS experiments indicated that MRS-N had the ability to oligomerize (Figs. S3A and S4A), suggesting that it induces the oligomerization of the M-complex and thus its potential to open a portal in the parasite plasma membrane. Alternatively, binding of exogenous tRNAs to the outer EMAPII-like domain of tRip could be a signal to dissociate and relocalize complex-bound aaRSs to parasite organelles and thus trigger their involvement in alternative functions. Our in vitro experiments do not support the dissociation of the complex upon tRNA binding (Fig. S5), but the outcome may be different in vivo. Also, unlike yeast or human MRSs, which exhibit nuclear localization (
      • Frechin M.
      • Enkler L.
      • Tetaud E.
      • Laporte D.
      • Senger B.
      • Blancard C.
      • et al.
      Expression of nuclear and mitochondrial genes encoding ATP synthase is synchronized by disassembly of a multisynthetase complex.
      ,
      • Ko Y.-G.
      • Kang Y.-S.
      • Kim E.-K.
      • Park S.G.
      • Kim S.
      Nucleolar localization of human methionyl–trna synthetase and its role in ribosomal RNA synthesis.
      ), P. falciparum MRS was not observed in the nucleus at any of the blood stages, even in the presence of growth factors (
      • Hussain T.
      • Yogavel M.
      • Sharma A.
      Inhibition of protein synthesis and malaria parasite development by drug targeting of methionyl-tRNA synthetases.
      ). Interestingly, these microscopy data also show that MRS is not homogeneously diffused in the schizonts’ cytosol, a situation consistent with a membrane-bound enzyme. Similarly, the indirect aminoacylation pathway of the apicoplast tRNAGln has been demonstrated (
      • Mailu B.M.
      • Li L.
      • Arthur J.
      • Nelson T.M.
      • Ramasamy G.
      • Fritz-Wolf K.
      • et al.
      Plasmodium apicoplast gln-tRNAGln biosynthesis utilizes a unique GatAB amidotransferase essential for erythrocytic stage parasites.
      ), suggesting that the nuclear-encoded QRS has no aminoacylation function in this organelle.
      tRip is an integral membrane protein that has been localized exclusively at the parasite surface at both the sporozoite stage (the infectious stage injected by mosquitoes into the vertebrate hosts) and the blood stage (
      • Bour T.
      • Mahmoudi N.
      • Kapps D.
      • Thiberge S.
      • Bargieri D.
      • Ménard R.
      • et al.
      Apicomplexa -specific tRip facilitates import of exogenous tRNAs into malaria parasites.
      ). To date, no other AIMP has been described with such localization, and, with the publication of the structure of the P. vivax tRip N-terminal domain, it appears that the transmembrane helix (α5) as we predicted (
      • Bour T.
      • Mahmoudi N.
      • Kapps D.
      • Thiberge S.
      • Bargieri D.
      • Ménard R.
      • et al.
      Apicomplexa -specific tRip facilitates import of exogenous tRNAs into malaria parasites.
      ) is deeply buried in the GST-like structure (
      • Gupta S.
      • Chhibber-Goel J.
      • Sharma M.
      • Parvez S.
      • Harlos K.
      • Sharma A.
      • et al.
      Crystal structures of the two domains that constitute the Plasmodium vivax p43 protein.
      ). Neither the crystal structure, nor the present study, have allowed identification of the transmembrane domain that anchors tRip in the parasite plasma membrane; however, the unique symmetry identified in the crystal structure of the P. vivax tRip-N dimer positions the C-terminal extremities of each monomer in the same direction, which is critical to orient both EMAPII-like domains toward the outside of the cell, while the rest of the complex is located inside the cell to allow the formation of the two MSCs. Interestingly, there are scattered elements in the literature that suggest the possibility for GST-like domains to traverse/interact/bind cellular membranes. Amongst them, chloride intracellular channels are a unique class of ion channels, which exist as both soluble and membrane bound oligomers (
      • Goodchild S.C.
      • Angstmann C.N.
      • Breit S.N.
      • Curmi P.M.G.
      • Brown L.J.
      Transmembrane extension and oligomerization of the CLIC1 chloride intracellular channel protein upon membrane interaction.
      ,
      • Hare J.E.
      • Goodchild S.C.
      • Breit S.N.
      • Curmi P.M.G.
      • Brown L.J.
      Interaction of human chloride intracellular channel protein 1 (CLIC1) with lipid bilayers: a fluorescence study.
      ). The ion channel hypothesis remains speculative. However, it has been demonstrated that chloride intracellular channels proteins have roles in diverse biological processes associated with membrane trafficking. They are often found associated with the actin cytoskeleton and to intracellular membranes (
      • Singh H.
      Two decades with dimorphic chloride intracellular channels (CLICs).
      ,
      • Argenzio E.
      • Moolenaar W.H.
      Emerging biological roles of Cl− intracellular channel proteins.
      ).
      tRip is not the only RNA-binding protein found on the surface of Plasmodium sporozoites: polyA-binding protein-1 (PABP-1) of Plasmodium yoelii, homologous to cytosolic PABPs, is localized on the surface of sporozoites (
      • Minns A.M.
      • Hart K.J.
      • Subramanian S.
      • Hafenstein S.
      • Lindner S.E.
      Nuclear, cytosolic, and surface-localized poly(A)-Binding proteins of Plasmodium yoelii.
      ). Similarly, several studies have discovered that the glycolytic GAPDH is surface localized at several stages of the P. falciparum life cycle; especially, it plays an important role in liver infection by sporozoites (
      • Lindner S.E.
      • Swearingen K.E.
      • Harupa A.
      • Vaughan A.M.
      • Sinnis P.
      • Moritz R.L.
      • et al.
      Total and putative surface proteomics of malaria parasite salivary gland sporozoites.
      ,
      • Cha S.-J.
      • Kim M.-S.
      • Pandey A.
      • Jacobs-Lorena M.
      Identification of GAPDH on the surface of Plasmodium sporozoites as a new candidate for targeting malaria liver invasion.
      ). Interestingly, among a variety of cellular processes, GAPDH has been implicated in the nuclear export of tRNAs, the stability and translation of mRNAs, and the replication and expression of several single-stranded RNA viruses (reviewed in (
      • White M.R.
      • Garcin E.D.
      The sweet side of RNA regulation: glyceraldehyde-3-phosphate dehydrogenase as a noncanonical RNA-binding protein: The sweet side of RNA regulation.
      )). Understanding why the parasite, particularly at the sporozoite stage, uses RNA-binding proteins on its surface may provide additional clues as to how it interacts with the host and/or other parasites.

      Experimental procedures

      Identification of tRip partners

      Sample preparation

      Four samples (biological replicates) of WT (GFPCON) blood stage of P. berghei parasites were assayed. Four negative control experiments were performed with the tRip-KO parasite (
      • Bour T.
      • Mahmoudi N.
      • Kapps D.
      • Thiberge S.
      • Bargieri D.
      • Ménard R.
      • et al.
      Apicomplexa -specific tRip facilitates import of exogenous tRNAs into malaria parasites.
      ) to define the background. Parasites were isolated from the blood of infected mice presenting parasitemia between 5 and 10%. The blood was centrifuged at room temperature for 10 min at 450 g to remove leukocytes. Pelleted erythrocytes were specifically lysed with different protocols: (a) by addition of erythrocyte lysis buffer (150 mM NH4Cl, 100 mM KHCO3, 10 mM EDTA, and 0.5% bovine serum albumin (w/v)) for 5 min on ice, (b) with saponin 0.1% (w/v) for 5 min on ice, and (c and d) by addition of activated streptolysin O (6000 U, 100 mM DTT) for 15 min at 37 °C. Samples were centrifuged for 8 min at 2000g and the supernatants were discarded. Pellets were resuspended in 1 ml PBS, and free parasites were separated from cellular debris by centrifugations (2 min at 100g) and transfer of the clean supernatant to a fresh tube. This operation was repeated four times until all contaminating cellular debris (especially intact red blood cells) were removed. Parasites were harvested by centrifugation for 8 min at 2000g and lysed in 1 ml cold extraction buffer (50 mM Tris–HCl pH 8.0, 150 mM NaCl, 1% Triton X-100, 10 mM DTT, and 1/50 protease inhibitor [cOmplete Tablets EDTA-free, Roche]) for 1 h on ice. Samples were mixed every 10 min, and the final cell lysate was centrifuged for 15 min at 10,000g to remove any insoluble material.

      Coimmunoprecipitation

      The coimmunoprecipitation (coIP) protocol was established by combining the kit protocol (MultiMACS Protein A/G MicroBeads kit; Miltenyi Biotec) with protocols described in (
      • Jones M.L.
      • Kitson E.L.
      • Rayner J.C.
      Plasmodium falciparum erythrocyte invasion: A conserved myosin associated complex.
      ,
      • Goel V.K.
      • Li X.
      • Chen H.
      • Liu S.-C.
      • Chishti A.H.
      • Oh S.S.
      Band 3 is a host receptor binding merozoite surface protein 1 during the Plasmodium falciparum invasion of erythrocytes.
      ). Briefly, parasite extracts (500 μg) were incubated with protein A-coated magnetic beads (50 μl/ml) for 30 min at 4 °C under gentle agitation. The cleared sample (1 ml) was then incubated with 6 μg of purified anti-tRip214-402 antibody and 50 μl of beads for 1 h under gentle agitation at 4 °C and loaded on a microcolumn applying a magnetic field. Successive washes (200 μl) were performed: twice with salt-free buffer (50 mM Tris–HCl pH 8.0, 1% NP-40), once with intermediate salt buffer (50 mM Tris–HCl pH 8.0, 150 mM NaCl, 1% NP-40), once with high salt buffer (50 mM Tris–HCl pH 8.0, 500 mM NaCl, 1% NP-40), and twice with salt-free buffer. Proteins bound to anti-tRip214-402 antibodies were eluted in 150 μl denaturing buffer (50 mM Tris–HCl pH 6.8, 50 mM DTT, 1% SDS, 0.025% bromophenol blue, and 10% glycerol) at 95 °C.

      MS and data analyses

      Eluted proteins were precipitated, reduced, and alkylated as described in (
      • Lange H.
      • Ndecky S.Y.A.
      • Gomez-Diaz C.
      • Pflieger D.
      • Butel N.
      • Zumsteg J.
      • et al.
      RST1 and RIPR connect the cytosolic RNA exosome to the Ski complex in Arabidopsis.
      ). Proteins were digested overnight with sequence-grade porcine trypsin (1:25, w/w; Promega), and generated peptides were analyzed on a NanoLC-2DPlus system (nanoFlexChiP module, Eksigent, AB Sciex) coupled to a TripleTOF 5600 mass spectrometer (AB Sciex) operating in positive mode. Peptides were loaded on a trap-and-eluate setup of C18 reverse-phase columns (ChiP C-18 precolumn 300 μm ID × 5 mm ChromXP and ChiP C-18 analytical column 75 μm ID × 15 cm ChromXP, Eksigent) in solvent A (0.1% formic acid in water) and separated with a 120 min gradient (5% to 40%) of solvent B (0.1% formic acid in acetonitrile) at a 300 nl/min flow rate. The mass spectrometer was operated in high-sensitivity data-dependent acquisition mode with Analyst software (v1.6, ABS ciex) on a 400 to 1250 m/z range. An external calibration was performed before each sample by monitoring 10 peptides of a β-galactosidase tryptic digest. A discovery “Top 20” method was used to select the most intense multiple-charged ions (2+ to 5+) for collision-induced dissociation fragmentation, with a cycle time of 3.3 s.
      MS data were searched simultaneously against the Mus musculus Swissprot database (release 2018_01, 33,744 entries) and P. berghei Uniprot database (release 2016_07, 32,170 entries) with Mascot algorithm (version 2.5, Matrix Science), using the software’s decoy strategy. Carbamidomethylation of cysteine was set as fixed modification; N-terminal protein acetylation, phosphorylation of serine/threonine/tyrosine, and oxidation of methionine were set as variable modifications; tryptic specificity with up to three miscleavages was used. The mass tolerances in MS and MS/MS were set to 30 ppm and 0.5 Da respectively. The resulting .dat files were imported into Proline 2.0 software (https://www.profiproteomics.fr/proline/) (
      • Bouyssié D.
      • Hesse A.-M.
      • Mouton-Barbosa E.
      • Rompais M.
      • Macron C.
      • Carapito C.
      • et al.
      Proline: an efficient and user-friendly software suite for large-scale proteomics.
      ) where they were validated with Mascot pretty rank equal to 1, 1% false discovery rate on both peptide spectrum matches, and protein sets (based on score). The total number of MS/MS fragmentation spectra was used to quantify each protein.
      Sample (b) displayed only 143 (WT) and 170 (KO) proteins due to the strong lysis conditions caused by the presence of saponin and was thus discarded for bioinformatics analysis. For the statistical analysis of the data, we compared the spectral count data collected for tRip samples against the KO control with a negative-binomial test and calculated the fold change, a p value, and an adjusted p value for each identified protein. Exploratory data analysis revealed the presence of batch effects in our experiments, and the edgeR negative binomial general linear model regression was run with the following design: null model; y∼batch and alternative model; y∼treatment + batch. For this study, considered protein partners are statistically enriched in the tRip samples with an adjusted p value < 0.05 and a minimum log fold change of 2. The MS proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE (
      • Perez-Riverol Y.
      • Bai J.
      • Bandla C.
      • García-Seisdedos D.
      • Hewapathirana S.
      • Kamatchinathan S.
      • et al.
      The PRIDE database resources in 2022: a hub for mass spectrometry-based proteomics evidences.
      ) partner repository with the dataset identifier PXD033057 and 10.6019/PXD033057.

      MS analysis of ERS-N, QRS-N, MRS-N, and tRip

      As for coIP analysis, proteins were digested with trypsin before LC-MS/MS analysis. Generated peptides were separated either with a 250 min gradient on the NanoLC-2DPlus-TripleTOF 5600 system using the same column setup as for coIP or with a 60 min gradient on the Easy-nanoLC-1000 system equipped with a C18 analytical column (75 μm ID × 25 cm nanoViper, 3 μm Acclaim PepMap, Thermo Fisher Scientific) and coupled to the Q-Exactive Plus mass spectrometer (Thermo Fisher Scientific) operating in data-dependent acquisition mode with Xcalibur software (Thermo Fisher) using a “Top10” strategy.
      Data were searched with Mascot algorithm (version 2.6, Matrix Science) on a combination of Escherichia coli sequences (Swissprot, E. coli K12, release 2022_01, 6507 entries) with contaminant proteins (release 2021_03, 111 entries) and the four sequences of interest. The mass tolerances in MS and MS/MS were set to 10 ppm and 0.02 Da, respectively. The resulting .dat files were imported into Proline 2.0 software where they were validated.

      Bioinformatics

      Protein sequences for tRip and cytosolic ERS, QRS, and MRS from all Plasmodium strains were retrieved from PlasmoDB (
      • PlasmoDB
      An integrative database of the Plasmodium falciparum genome. Tools for accessing and analyzing finished and unfinished sequence data.
      ). Related proteins from other organisms were searched using NCBI BLAST and relevant sequences were fetched from Uniprot (
      • Bateman A.
      • Martin M.-J.
      • Orchard S.
      • Magrane M.
      • Agivetova R.
      • Ahmad S.
      • et al.
      The UniProt Consortium
      UniProt: The universal protein knowledgebase in 2021.
      ). All MSAs were performed using T-coffee (https://tcoffee.crg.eu/) (
      • Magis C.
      • Taly J.-F.
      • Bussotti G.
      • Chang J.-M.
      • Di Tommaso P.
      • Erb I.
      • et al.
      T-Coffee: tree-based consistency objective function for alignment evaluation. In multiple sequence alignment methods.
      ). LCRs in Plasmodium proteins were detected by aligning the sequences of different strains. Delimitation of additional sequences appended to Plasmodium aaRSs was achieved by alignment against prokaryotic homologs. HHpred (
      • Soding J.
      • Biegert A.
      • Lupas A.N.
      The HHpred interactive server for protein homology detection and structure prediction.
      ) and Batch CDD search (
      • Marchler-Bauer A.
      • Derbyshire M.K.
      • Gonzales N.R.
      • Lu S.
      • Chitsaz F.
      • Geer L.Y.
      • et al.
      Cdd: NCBI’s conserved domain database.
      ) were used to identify conserved domains in Plasmodium proteins by submitting either full-length sequences or specific segments. Secondary structure predictions were carried out with Quick 2D (
      • Zimmermann L.
      • Stephens A.
      • Nam S.-Z.
      • Rau D.
      • Kübler J.
      • Lozajic M.
      • et al.
      A completely reimplemented MPI bioinformatics toolkit with a new HHpred server at its core.
      ). Template-based 3D models of P. berghei proteins were generated using the Raptor X web server (
      • Källberg M.
      • Wang H.
      • Wang S.
      • Peng J.
      • Wang Z.
      • Lu H.
      • et al.
      Template-based protein structure modeling using the RaptorX web server.
      ).

      Plasmid constructions and production of recombinant proteins

      Synthetic genes (GenScript) encoding P. berghei tRip and the GST-like domains of ERS, QRS, and MRS were cloned into pET15b. tRip was fused to an N-terminal 6His tag. The N-terminal domains of ERS, QRS, and MRS were fused either to a removable C-terminal 6His tag (removed by thrombin cleavage) or a removable SUMO-6His tag (removed by TEV cleavage). Besides improving solubility, addition of the SUMO peptide resulted in proteins of different sizes that could be distinguished by SDS-PAGE. Proteins with no tag were obtained by introducing a stop codon (TAA) before the thrombin site and mutations at specific amino acid positions were generated using the QuickChange site-directed mutagenesis kit (Agilent). Since the AT content of most Plasmodium protein genes is above 70% (
      • Gardner M.J.
      • Hall N.
      • Fung E.
      • White O.
      • Berriman M.
      • Hyman R.W.
      • et al.
      Genome sequence of the human malaria parasite Plasmodium falciparum.
      ), sequences were optimized to increase expression of tRip and the three aaRSs. Sequences were adapted to human codon usage, even though the proteins were expressed in E. coli. In this way, the AT content was decreased, while some rare codons were retained to slow down translation of the recombinant protein in E. coli. All cultures were started from freshly transformed E. coli BL21(DE3) cells in LB medium supplemented with 100 μg/ml ampicillin. Cultures were inoculated to a starting A600nm of 0.015 and incubated at 30 °C and 180 rpm until an A600nm of 0.8 was reached. Protein expression was induced overnight at 16 °C with 0.5 mM isopropyl-β-D-thiogalactopyranoside. Induction of protein expression at low temperature (16 °C) was crucial to obtain soluble proteins. Expression and solubility of recombinant proteins were checked by SDS-PAGE.

      In vitro pull-down assays

      Interactions between tRip and the three other GST-like domains were investigated by in vitro pull-down assays. Bacteria expressing a 6His-tagged protein (bait) as well as one or more bacteria expressing untagged proteins (prey) were suspended in 1.2 ml binding buffer (50 mM Hepes–NaOH pH 8.0, 300 mM NaCl, 10% (v/v) glycerol, 20 mM imidazole, and 5 mM 2-mercaptoethanol [β-Me]) supplemented with 0.005% (w/v) n-dodecyl β-D-maltoside (DDM). Cells were disrupted by sonicating twice for 10 s at amplitude 40 (Vibracell 75022 UltraSonic Processor), and the crude extract was centrifuged at 15,000g for 15 min at 4 °C. The protein extract was incubated with 75 μl Ni-iminodiacetic acid agarose (Bio-Rad) for 30 min at room temperature on a tube rotator, and the mixture was transferred to an empty chromatography column (Sigma–Aldrich). The resin was washed twice with 2.5 ml of binding buffer with DDM and once with 2.5 ml of binding buffer without DDM. Proteins specifically bound to the resin were eluted with 100 μl of binding buffer containing 250 mM imidazole. The protein contents of both the centrifuged extract and the eluate were analyzed by SDS-PAGE. To estimate the abundance of proteins in the eluate, the density of each band was quantified using Image J (https://imagej.nih.gov/ij/) (
      • Abràmoff M.,D.
      • Magalhães P.,J.
      • Ram S.,J.
      Image processing with ImageJ.
      ). The value was divided by the number of arginine, lysine, and histidine residues to account for differences in Coomassie staining. Error bars were calculated from at least three replicates. Because the solubilities of WT and mutated proteins varied, the volumes of bacterial cultures were adapted to have similar amounts of each protein in the initial mixture.

      Purification of complexes

      Bacteria expressing individual recombinant proteins (one bait and several preys) were lysed together in lysis buffer (50 mM Hepes–NaOH pH 8.0, 300 mM NaCl, 10% (v/v) glycerol, 5 mM β-Me, and 0.005% (w/v) DDM). The volume was adapted so that 500 ml of culture was resuspended in 25 ml of lysis buffer. Each 25 ml sample was sonicated in ice for 7 min at 120 V (Annemasse Ultrasons apparatus) and ultracentrifuged at 45,000g for 45 min at 4 °C (Optima XE-90, Beckman-Coulter). The clarified extract was loaded onto a 1 ml Ni-NTA (His-Select HF, Sigma–Aldrich) column equilibrated with lysis buffer (BioLogic DuoFlow chromatography system). The column was washed with 15 ml of lysis buffer supplemented with 20 mM imidazole, 6 ml of a linear gradient (0.3–1.0 M NaCl) in the same buffer, 6 ml of the inverted gradient back to 0.3 M, and 33 ml of lysis buffer containing 20 mM imidazole. Retained proteins were eluted with 25 ml of lysis buffer containing 250 mM imidazole and the collected fractions for analysis by SDS-PAGE. Fractions were pooled and concentrated (Amicon ultrafiltration devices, Merck Millipore) and further analyzed on a SepFast 10/300 6 to 5000 kDa SEC column (BioToolomics) at 0.2 ml/min in SEC buffer (25 mM Hepes–NaOH pH 7.0, 300 mM NaCl, 5% (v/v) glycerol, 0.005% (w/v) DDM, and 5 mM β-Me). Fractions were analyzed on SDS-PAGE, and the abundance of each protein partner was estimated as described previously. Protein concentration was determined with a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific) and the A260/A280 ratio was used to assess nucleic acid contamination (A260/A280 ≈ 0.5 for pure protein samples). SEC column was calibrated using either the MWGF1000 kit (Sigma–Aldrich) or the gel filtration standard (Bio-Rad). If removal of SUMO tags was required, proteins eluted from the Ni-NTA were dialyzed overnight at 4 °C against lysis buffer in the presence of TEV protease (6His-tagged, 1 μg of TEV per 25 μg of protein; homemade). The SUMO tag, 6His-TEV, and noncleaved complexes were eliminated by running the sample onto a second Ni-NTA column.

      Light-scattering measurements

      DLS and SLS measurements were performed with a Wyatt Technology DynaPro Nanostar instrument (100 mW He-Ne laser, λ = 658 nm, DLS θ = 90°, SLS θ = 90°, 500 channel correlator) using a 1 μl quartz cuvette (serial number JC-164) calibrated with toluene at 25 °C. Ten measurements, each one composed of 10 acquisitions of 5 s, were recorded at 20 °C and processed with DYNAMICS, version 7.8.1.3. (Wyatt Technology) Solvent refractive index and absolute viscosity were estimated using Malvern Panalytical DLS software, version 8, and the particle increment of refractive index (dn/dc) was assumed to be 0.185 ml/g. The intensity of the solvent alone (filtered through a membrane with a pore diameter of 0.1 μm) was measured and subtracted from the sample in SLS calculations. Several dilutions of freshly purified protein samples in SEC buffer were ultracentrifuged for 1 h at 100,000g at 4 °C (S45A rotor in Sorvall Hitachi Discovery M150E microultracentrifuge) before light-scattering measurements. Sample concentrations were verified using a NanoDrop ND-1000 spectrophotometer.

      SEC-MALS experiments

      SEC-MALS was performed at 25 °C using a Superose 6 10/300 GL column (GE Healthcare) attached to a light-scattering detector (miniDAWN TREOS, Wyatt Technology) and a differential refractive index detector (Optilab T-rEX, Wyatt Technology). The column was equilibrated with SEC buffer and detector normalization was achieved with bovine serum albumin (Sigma–Aldrich). Samples were ultracentrifuged 1 h at 100,000g at 4 °C before injection of Q-complex (30 μl at 6.8 mg/ml) or M-complex (50 μl at 8.6 mg/ml). Data were analyzed using ASTRA (Wyatt Technology), version 6.1.7.17.

      EMSAs

      Agarose gels

      Protein complexes (3.3 μM) were incubated with increasing concentrations of total yeast tRNA (0.55–6.6 μM) in binding buffer (25 mM Hepes–NaOH pH 7.0, 150 mM NaCl, 10% (v/v) glycerol, 5 mM MgCl2, 0.005% (w/v) DDM, and 5 mM β-Me), and 6.7 μM of dodeca-dT (competitor) for 20 min on ice. tRNA binding was analyzed by electrophoresis (75 V, 1 h 30 at 4 °C) on a 1% (w/v) low-melting agarose gel (Quantum Biotechnologies) in Tris–borate–EDTA buffer. Gels were first stained with ethidium bromide and then with InstantBlue Coomasie protein stain (Expedeon Ltd). Bands containing protein–RNA complexes were excised, melted at 95 °C, and mixed with one volume of 2X SDS-PAGE loading buffer. The pH was adjusted by adding a few microliters of concentrated NaOH, and samples were analyzed on 1.5 mm thick 12% SDS-PAGE.

      Polyacrylamide affinity coelectrophoresis

      Increasing concentrations of peptides (62.5 nM to 1 μM) were embedded in a 1.5 × 80 × 100 mm3 6% (19/1) polyacrylamide gel. Samples (5 μl, 5 nM [3′-32P] RNA in 25 mM Hepes–NaOH pH 7.0, 5 mM MgCl2, 10% glycerol, and bromophenol blue) of radiolabeled total yeast tRNA were electrophoresed for 90 min 70 V at 4 °C. Gels were dried and analyzed on a Phosphor-Imager (Typhoon FLA 7000).

      Data availability

      All data files are available in the article and supplementary data. The mass spectrometry proteomics data are included in Table S1 and have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier PXD033057 and 10.6019/PXD033057.

      Supporting information

      This article contains supporting information.

      Conflict of interest

      The authors declare that they have no conflicts of interest with the contents of this article.

      Acknowledgments

      We are grateful to Philippe Hammann, Lauriane Kuhn and Béatrice Chane Woon Ming for LC-MS/MS analysis, Pierre Poussin-Courmontagne for SEC-MALS experiments, and to Prof Tamara Hendrickson for providing comments on this article. This work was performed under the framework of the Interdisciplinary Thematic Institute IMCBio, as part of the ITI 2021 to 2028 program of the University of Strasbourg, CNRS and Inserm.

      Author contributions

      M. F. methodology; J. R. J. P., D. K., J. C, and C. P investigation; J. R. J. P. and M. F. data curation; M. F. and J. R. J. P. writing–original draft; M. F. and J. R. J. P. writing–review & editing; M. F. funding acquisition.

      Funding and additional information

      This work was performed under the framework of the Interdisciplinary Thematic Institute IMCBio, as part of the ITI 2021-2028 program of the University of Strasbourg, CNRS and Inserm. It was supported by IdEx Unistra (ANR-10-IDEX-0002), by SFRI-STRAT’US project (ANR 20-SFRI-0012), and EUR IMCBio (IMCBio ANR-17-EURE-0023) under the framework of the French Investments for the Future Program », by previous Labex NetRNA (ANR-10-LABX-0036), by the CNRS and the Université de Strasbourg, IdEx “Equipement mi-lourd” (2015), and CONACYT- Mexico (grant number 439648) to JRJP.

      References

        • Rubio Gomez M.A.
        • Ibba M.
        Aminoacyl-tRNA synthetases.
        RNA. 2020; 26: 910-936
        • Park S.G.
        • Ewalt K.L.
        • Kim S.
        Functional expansion of aminoacyl-tRNA synthetases and their interacting factors: New perspectives on housekeepers.
        Trends Biochem. Sci. 2005; 30: 569-574
        • Pang Y.L.J.
        • Poruri K.
        • Martinis S.A.
        tRNA synthetase: tRNA aminoacylation and beyond.
        Wiley Interdiscip. Rev. RNA. 2014; 5: 461-480
        • Guo M.
        • Schimmel P.
        Essential nontranslational functions of tRNA synthetases.
        Nat. Chem. Biol. 2013; 9: 145-153
        • Hyeon D.Y.
        • Kim J.H.
        • Ahn T.J.
        • Cho Y.
        • Hwang D.
        • Kim S.
        Evolution of the multi-tRNA synthetase complex and its role in cancer.
        J. Biol. Chem. 2019; 294: 5340-5351
        • Han J.M.
        • Kim J.Y.
        • Kim S.
        Molecular network and functional implications of macromolecular tRNA synthetase complex.
        Biochem. Biophys. Res. Commun. 2003; 303: 985-993
        • Lee S.W.
        • Cho B.H.
        • Park S.G.
        • Kim S.
        Aminoacyl-tRNA synthetase complexes: Beyond translation.
        J. Cell Sci. 2004; 117: 3725-3734
        • Cui H.
        • Kapur M.
        • Diedrich J.K.
        • Yates J.R.
        • Ackerman S.L.
        • Schimmel P.
        Regulation of ex-translational activities is the primary function of the multi-tRNA synthetase complex.
        Nucl. Acids Res. 2021; 49: 3603-3616
        • Guo M.
        • Yang X.-L.
        Architecture and metamorphosis. In aminoacyl-tRNA synthetases in biology and medicine.
        in: Topics in Current Chemistry. 344. Springer Netherlands, Dordrecht2013: 89-118
        • Havrylenko S.
        • Mirande M.
        Aminoacyl-tRNA synthetase complexes in evolution.
        IJMS. 2015; 16: 6571-6594
        • Karanasios E.
        • Simos G.
        Building arks for tRNA: Structure and function of the Arc1p family of non-catalytic tRNA-binding proteins.
        FEBS Lett. 2010; 584: 3842-3849
        • Kapps D.
        • Cela M.
        • Théobald-Dietrich A.
        • Hendrickson T.
        • Frugier M.
        OB or Not OB: Idiosyncratic utilization of the tRNA-binding OB-fold domain in unicellular, pathogenic eukaryotes.
        FEBS Lett. 2016; 590: 4180-4191
        • Simos G.
        • Segref A.
        • Fasiolo F.
        • Hellmuth K.
        • Shevchenko A.
        • Mann M.
        • et al.
        The yeast protein Arc1p binds to tRNA and functions as a cofactor for the methionyl- and glutamyl-tRNA synthetases.
        EMBO J. 1996; 15: 5437-5448
        • van Rooyen J.M.
        • Murat J.-B.
        • Hammoudi P.-M.
        • Kieffer-Jaquinod S.
        • Coute Y.
        • Sharma A.
        • et al.
        Assembly of the novel five-component apicomplexan multi-aminoacyl-tRNA synthetase complex is driven by the hybrid scaffold protein tg-p43.
        PLoS One. 2014; 9e89487
        • Cestari I.
        • Kalidas S.
        • Monnerat S.
        • Anupama A.
        • Phillips M.A.
        • Stuart K.
        A multiple aminoacyl-tRNA synthetase complex that enhances tRNA-aminoacylation in african trypanosomes.
        Mol. Cell Biol. 2013; 33: 4872-4888
        • Ahn H.-C.
        • Kim S.
        • Lee B.-J.
        Solution structure and p43 binding of the p38 leucine zipper motif: coiled-coil interactions mediate the association between p38 and p43.
        FEBS Lett. 2003; 542: 119-124
        • Havrylenko S.
        • Legouis R.
        • Negrutskii B.
        • Mirande M.
        Caenorhabditis elegans evolves a new architecture for the multi-aminoacyl-tRNA synthetase complex.
        J. Biol. Chem. 2011; 286: 28476-28487
        • Negrutskii B.S.
        • Deutscher M.P.
        Channeling of aminoacyl-tRNA for protein synthesis in vivo.
        Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 4991-4995
        • Simos G.
        • Sauer A.
        • Fasiolo F.
        • Hurt E.C.
        A conserved domain within Arc1p delivers tRNA to aminoacyl-tRNA synthetases.
        Mol. Cell. 1998; 1: 235-242
        • Kyriacou S.V.
        • Deutscher M.P.
        An important role for the multienzyme aminoacyl-tRNA synthetase complex in mammalian translation and cell growth.
        Mol. Cell. 2008; 29: 419-427
        • Golinelli-Cohen M.-P.
        • Mirande M.
        Arc1p is required for cytoplasmic confinement of synthetases and tRNA.
        Mol. Cell Biochem. 2007; 300: 47-59
        • Arif A.
        • Jia J.
        • Mukhopadhyay R.
        • Willard B.
        • Kinter M.
        • Fox P.L.
        Two-site phosphorylation of EPRS coordinates multimodal regulation of noncanonical translational control activity.
        Mol. Cell. 2009; 35: 164-180
        • Frechin M.
        • Enkler L.
        • Tetaud E.
        • Laporte D.
        • Senger B.
        • Blancard C.
        • et al.
        Expression of nuclear and mitochondrial genes encoding ATP synthase is synchronized by disassembly of a multisynthetase complex.
        Mol. Cell. 2014; 56: 763-776
        • Han J.M.
        • Lee M.J.
        • Park S.G.
        • Lee S.H.
        • Razin E.
        • Choi E.-C.
        • et al.
        Hierarchical network between the components of the multi-tRNA synthetase complex.
        J. Biol. Chem. 2006; 281: 38663-38667
        • Laporte D.
        • Huot J.L.
        • Bader G.
        • Enkler L.
        • Senger B.
        • Becker H.D.
        Exploring the evolutionary diversity and assembly modes of multi-aminoacyl-tRNA synthetase complexes: Lessons from unicellular organisms.
        FEBS Lett. 2014; 588: 4268-4278
        • Koonin E.V.
        • Tatusov R.L.
        • Altschul S.F.
        • Bryant S.H.
        • Mushegian A.R.
        • Bork P.
        • et al.
        Eukaryotic translation elongation factor 1γ contains a glutathione transferase domain-Study of a diverse, ancient protein super family using motif search and structural modeling.
        Protein Sci. 1994; 3: 2045-2055
        • Bec G.
        • Kerjan P.
        • Waller J.P.
        Reconstitution in vitro of the valyl-tRNA synthetase-elongation factor (EF) 1 beta gamma delta complex. essential roles of the NH2-terminal extension of valyl-tRNA synthetase and of the EF-1 delta subunit in complex formation.
        J. Biol. Chem. 1994; 269: 2086-2092
        • Guo M.
        • Yang X.L.
        • Schimmel P.
        New functions of aminoacyl-tRNA synthetases beyond translation.
        Nat. Rev. Mol. Cell Biol. 2010; 11: 668-674
        • Achilonu I.
        • Elebo N.
        • Hlabano B.
        • Owen G.R.
        • Papathanasopoulos M.
        • Dirr H.W.
        An update on the biophysical character of the human eukaryotic elongation factor 1 beta: Perspectives from interaction with elongation factor 1 gamma.
        J. Mol. Recognit. 2018; 31e2708
        • Koehler C.
        • Round A.
        • Simader H.
        • Suck D.
        • Svergun D.
        Quaternary structure of the yeast Arc1p-aminoacyl-tRNA synthetase complex in solution and its compaction upon binding of tRNAs.
        Nucl. Acids Res. 2013; 41: 667-676
        • Kaminska M.
        • Havrylenko S.
        • Decottignies P.
        • Gillet S.
        • Maréchal P.L.
        • Negrutskii B.
        • et al.
        Dissection of the structural organization of the aminoacyl-tRNA synthetase complex.
        J. Biol. Chem. 2009; 284: 6053-6060
        • Mirande M.
        The aminoacyl-tRNA synthetase complex. In macromolecular protein complexes.
        in: Subcellular Biochemistry. 83. Springer International Publishing, Cham2017: 505-522
        • Khan K.
        • Baleanu-Gogonea C.
        • Willard B.
        • Gogonea V.
        • Fox P.L.
        3-Dimensional architecture of the human multi-tRNA synthetase complex.
        Nucl. Acids Res. 2020; 48: 8740-8754
        • Dirr H.
        • Reinemer P.
        • Huber R.
        X-ray crystal structures of cytosolic glutathione S-transferases. Implications for protein architecture, substrate recognition and catalytic function.
        Eur. J. Biochem. 1994; 220: 645-661
        • Armstrong R.N.
        Structure, catalytic mechanism, and evolution of the glutathione transferases.
        Chem. Res. Toxicol. 1997; 10: 2-18
        • Aceto A.
        • Dragani B.
        • Melino S.
        • Allocati N.
        • Masulli M.
        • Ilio C.D.
        • et al.
        Identification of an N-capping box that affects the α6-helix propensity in glutathione S-transferase superfamily proteins: a role for an invariant aspartic residue.
        Biochem. J. 1997; 322: 229-234
        • Cocco R.
        • Stenberg G.
        • Dragani B.
        • Principe D.R.
        • Paludi D.
        • Mannervik B.
        • et al.
        The folding and stability of human alpha class glutathione transferase A1-1 depend on distinct roles of a conserved N-capping box and hydrophobic staple motif.
        J. Biol. Chem. 2001; 276: 32177-32183
        • Simader H.
        • Hothorn M.
        • Köhler C.
        • Basquin J.
        • Simos G.
        • Suck D.
        Structural basis of yeast aminoacyl-tRNA synthetase complex formation revealed by crystal structures of two binary sub-complexes.
        Nucl. Acids Res. 2006; 34: 3968-3979
        • Karanasios E.
        • Simader H.
        • Panayotou G.
        • Suck D.
        • Simos G.
        Molecular determinants of the yeast arc1p–aminoacyl-tRNA synthetase complex assembly.
        J. Mol. Biol. 2007; 374: 1077-1090
        • Hahn H.
        • Park S.H.
        • Kim H.-J.
        • Kim S.
        • Han B.W.
        The DRS–AIMP2–EPRS subcomplex acts as a pivot in the multi-tRNA synthetase complex.
        IUCrJ. 2019; 6: 958-967
        • Cho H.Y.
        • Lee H.J.
        • Choi Y.S.
        • Kim D.K.
        • Jin K.S.
        • Kim S.
        • et al.
        Symmetric assembly of a decameric subcomplex in human multi-tRNA synthetase complex via interactions between glutathione transferase-homology domains and aspartyl-tRNA synthetase.
        J. Mol. Biol. 2019; 431: 4475-4496
        • Cho H.Y.
        • Maeng S.J.
        • Cho H.J.
        • Choi Y.S.
        • Chung J.M.
        • Lee S.
        • et al.
        Assembly of multi-tRNA synthetase complex via heterotetrameric glutathione transferase-homology domains.
        J. Biol. Chem. 2015; 290: 29313-29328
        • Bour T.
        • Mahmoudi N.
        • Kapps D.
        • Thiberge S.
        • Bargieri D.
        • Ménard R.
        • et al.
        Apicomplexa -specific tRip facilitates import of exogenous tRNAs into malaria parasites.
        Proc. Natl. Acad. Sci. U. S. A. 2016; 113: 4717-4722
        • Gupta S.
        • Chhibber-Goel J.
        • Sharma M.
        • Parvez S.
        • Harlos K.
        • Sharma A.
        • et al.
        Crystal structures of the two domains that constitute the Plasmodium vivax p43 protein.
        Acta Crystallogr. D Struct. Biol. 2020; 76: 135-146
        • Frugier M.
        • Bour T.
        • Ayach M.
        • Santos M.A.
        • Rudinger-Thirion J.
        • Theobald-Dietrich A.
        • et al.
        Low Complexity Regions behave as tRNA sponges to help co-translational folding of plasmodial proteins.
        FEBS Lett. 2010; 584: 448-454
        • Chaudhry S.R.
        • Lwin N.
        • Phelan D.
        • Escalante A.A.
        • Battistuzzi F.U.
        Comparative analysis of low complexity regions in Plasmodia.
        Sci. Rep. 2018; 8: 335
        • Singh G.P.
        • Chandra B.R.
        • Bhattacharya A.
        • Akhouri R.R.
        • Singh S.K.
        • Sharma A.
        Hyper-expansion of asparagines correlates with an abundance of proteins with prion-like domains in Plasmodium falciparum.
        Mol. Biochem. Parasitol. 2004; 137: 307-319
        • Källberg M.
        • Wang H.
        • Wang S.
        • Peng J.
        • Wang Z.
        • Lu H.
        • et al.
        Template-based protein structure modeling using the RaptorX web server.
        Nat. Protoc. 2012; 7: 1511-1522
        • Frugier M.
        • Moulinier L.
        • Giegé R.
        A domain in the N-terminal extension of class IIb eukaryotic aminoacyl-tRNA synthetases is important for tRNA binding.
        EMBO J. 2000; 19: 2371-2380
        • Francin M.
        • Mirande M.
        Functional dissection of the eukaryotic-specific tRNA-interacting factor of lysyl-tRNA synthetase.
        J. Biol. Chem. 2003; 278: 1472-1479
        • Crepin T.
        • Peterson F.
        • Haertlein M.
        • Jensen D.
        • Wang C.
        • Cusack S.
        • et al.
        A hybrid structural model of the complete Brugia malayi cytoplasmic asparaginyl-tRNA synthetase.
        J. Mol. Biol. 2011; 405: 1056-1069
        • Marchler-Bauer A.
        • Bo Y.
        • Han L.
        • He J.
        • Lanczycki C.J.
        • Lu S.
        • et al.
        CDD/SPARCLE: Functional classification of proteins via subfamily domain architectures.
        Nucl. Acids Res. 2017; 45: D200-D203
        • Galani K.
        • Großhans H.
        • Deinert K.
        • Hurt E.C.
        • Simos G.
        The intracellular location of two aminoacyl-tRNA synthetases depends on complex formation with Arc1p.
        EMBO J. 2001; 20: 6889-6898
        • Deinert K.
        • Fasiolo F.
        • Hurt E.C.
        • Simos G.
        Arc1p organizes the yeast aminoacyl-tRNA synthetase complex and stabilizes its interaction with the cognate tRNAs.
        J. Biol. Chem. 2001; 276: 6000-6008
        • Quevillon S.
        • Agou F.
        • Robinson J.-C.
        • Mirande M.
        The p43 component of the mammalian multi-synthetase complex is likely to Be the precursor of the endothelial monocyte-activating polypeptide II cytokine.
        J. Biol. Chem. 1997; 272: 32573-32579
        • Kaminska M.
        • Deniziak M.
        • Kerjan P.
        • Barciszewski J.
        • Mirande M.
        A recurrent general RNA binding domain appended to plant methionyl-tRNA synthetase acts as a cis-acting cofactor for aminoacylation.
        EMBO J. 2000; 19: 6908-6917
        • Cahuzac B.
        A recurrent RNA-binding domain is appended to eukaryotic aminoacyl-tRNA synthetases.
        EMBO J. 2000; 19: 445-452
        • Cela M.
        • Théobald-Dietrich A.
        • Rudinger-Thirion J.
        • Wolff P.
        • Geslain R.
        • Frugier M.
        Identification of host tRNAs preferentially recognized by the Plasmodium surface protein tRip.
        Nucl. Acids Res. 2021; 49: 10618-10629
        • Davies H.M.
        • Nofal S.D.
        • McLaughlin E.J.
        • Osborne A.R.
        Repetitive sequences in malaria parasite proteins.
        FEMS Microbiol. Rev. 2017; 41: 923-940
        • Ko Y.-G.
        • Kang Y.-S.
        • Kim E.-K.
        • Park S.G.
        • Kim S.
        Nucleolar localization of human methionyl–trna synthetase and its role in ribosomal RNA synthesis.
        J. Cell Biol. 2000; 149: 567-574
        • Hussain T.
        • Yogavel M.
        • Sharma A.
        Inhibition of protein synthesis and malaria parasite development by drug targeting of methionyl-tRNA synthetases.
        Antimicrob. Agents Chemother. 2015; 59: 1856-1867
        • Mailu B.M.
        • Li L.
        • Arthur J.
        • Nelson T.M.
        • Ramasamy G.
        • Fritz-Wolf K.
        • et al.
        Plasmodium apicoplast gln-tRNAGln biosynthesis utilizes a unique GatAB amidotransferase essential for erythrocytic stage parasites.
        J. Biol. Chem. 2015; 290: 29629-29641
        • Goodchild S.C.
        • Angstmann C.N.
        • Breit S.N.
        • Curmi P.M.G.
        • Brown L.J.
        Transmembrane extension and oligomerization of the CLIC1 chloride intracellular channel protein upon membrane interaction.
        Biochemistry. 2011; 50: 10887-10897
        • Hare J.E.
        • Goodchild S.C.
        • Breit S.N.
        • Curmi P.M.G.
        • Brown L.J.
        Interaction of human chloride intracellular channel protein 1 (CLIC1) with lipid bilayers: a fluorescence study.
        Biochemistry. 2016; 55: 3825-3833
        • Singh H.
        Two decades with dimorphic chloride intracellular channels (CLICs).
        FEBS Lett. 2010; 584: 2112-2121
        • Argenzio E.
        • Moolenaar W.H.
        Emerging biological roles of Cl− intracellular channel proteins.
        J. Cell Sci. 2016; 129: 4165-4174
        • Minns A.M.
        • Hart K.J.
        • Subramanian S.
        • Hafenstein S.
        • Lindner S.E.
        Nuclear, cytosolic, and surface-localized poly(A)-Binding proteins of Plasmodium yoelii.
        mSphere. 2018; 3e00435-17
        • Lindner S.E.
        • Swearingen K.E.
        • Harupa A.
        • Vaughan A.M.
        • Sinnis P.
        • Moritz R.L.
        • et al.
        Total and putative surface proteomics of malaria parasite salivary gland sporozoites.
        Mol. Cell Proteomics. 2013; 12: 1127-1143
        • Cha S.-J.
        • Kim M.-S.
        • Pandey A.
        • Jacobs-Lorena M.
        Identification of GAPDH on the surface of Plasmodium sporozoites as a new candidate for targeting malaria liver invasion.
        J. Exp. Med. 2016; 213: 2099-2112
        • White M.R.
        • Garcin E.D.
        The sweet side of RNA regulation: glyceraldehyde-3-phosphate dehydrogenase as a noncanonical RNA-binding protein: The sweet side of RNA regulation.
        WIREs RNA. 2016; 7: 53-70
        • Jones M.L.
        • Kitson E.L.
        • Rayner J.C.
        Plasmodium falciparum erythrocyte invasion: A conserved myosin associated complex.
        Mol. Biochem. Parasitol. 2006; 147: 74-84
        • Goel V.K.
        • Li X.
        • Chen H.
        • Liu S.-C.
        • Chishti A.H.
        • Oh S.S.
        Band 3 is a host receptor binding merozoite surface protein 1 during the Plasmodium falciparum invasion of erythrocytes.
        Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 5164-5169
        • Lange H.
        • Ndecky S.Y.A.
        • Gomez-Diaz C.
        • Pflieger D.
        • Butel N.
        • Zumsteg J.
        • et al.
        RST1 and RIPR connect the cytosolic RNA exosome to the Ski complex in Arabidopsis.
        Nat Commun. 2019; 10: 3871-3882
        • Bouyssié D.
        • Hesse A.-M.
        • Mouton-Barbosa E.
        • Rompais M.
        • Macron C.
        • Carapito C.
        • et al.
        Proline: an efficient and user-friendly software suite for large-scale proteomics.
        Bioinformatics. 2020; 36: 3148-3155
        • Perez-Riverol Y.
        • Bai J.
        • Bandla C.
        • García-Seisdedos D.
        • Hewapathirana S.
        • Kamatchinathan S.
        • et al.
        The PRIDE database resources in 2022: a hub for mass spectrometry-based proteomics evidences.
        Nucl. Acids Res. 2022; 50: D543-D552
        • PlasmoDB
        An integrative database of the Plasmodium falciparum genome. Tools for accessing and analyzing finished and unfinished sequence data.
        Nucl. Acids Res. 2001; 29: 66-69
        • Bateman A.
        • Martin M.-J.
        • Orchard S.
        • Magrane M.
        • Agivetova R.
        • Ahmad S.
        • et al.
        • The UniProt Consortium
        UniProt: The universal protein knowledgebase in 2021.
        Nucl. Acids Res. 2021; 49: D480-D489
        • Magis C.
        • Taly J.-F.
        • Bussotti G.
        • Chang J.-M.
        • Di Tommaso P.
        • Erb I.
        • et al.
        T-Coffee: tree-based consistency objective function for alignment evaluation. In multiple sequence alignment methods.
        in: Methods in Molecular Biology. 1079. Humana Press, Totowa, NJ2014: 117-129
        • Soding J.
        • Biegert A.
        • Lupas A.N.
        The HHpred interactive server for protein homology detection and structure prediction.
        Nucl. Acids Res. 2005; 33: W244-W248
        • Marchler-Bauer A.
        • Derbyshire M.K.
        • Gonzales N.R.
        • Lu S.
        • Chitsaz F.
        • Geer L.Y.
        • et al.
        Cdd: NCBI’s conserved domain database.
        Nucl. Acids Res. 2015; 43: D222-D226
        • Zimmermann L.
        • Stephens A.
        • Nam S.-Z.
        • Rau D.
        • Kübler J.
        • Lozajic M.
        • et al.
        A completely reimplemented MPI bioinformatics toolkit with a new HHpred server at its core.
        J. Mol. Biol. 2018; 430: 2237-2243
        • Gardner M.J.
        • Hall N.
        • Fung E.
        • White O.
        • Berriman M.
        • Hyman R.W.
        • et al.
        Genome sequence of the human malaria parasite Plasmodium falciparum.
        Nature. 2002; 419: 498-511
        • Abràmoff M.,D.
        • Magalhães P.,J.
        • Ram S.,J.
        Image processing with ImageJ.
        Biophot. Int. 2004; 11: 36-42
        • Robinson J.-C.
        • Kerjan P.
        • Mirande M.
        Macromolecular assemblage of aminoacyl-tRNA synthetases: quantitative analysis of protein-protein interactions and mechanism of complex assembly.
        J. Mol. Biol. 2000; 304: 983-994