LeuRS (leucyl-tRNA synthetase) catalyses the esterification of tRNAsLeu with leucine. This family of enzymes is divided into prokaryotic and eukaryal/archaeal groups according to the presence and position of specific insertions and extensions. In the present study, we investigated the function of LSD1 (leucine-specific domain 1), which is naturally present in eukaryal/archaeal LeuRSs, but absent from prokaryotic LeuRSs. When mutated in their common domain, the eukaryal and archaeal LeuRSs exhibited defects in the first reaction step of amino acid activation with variations of leucine or ATP-binding strength, whereas the tRNA aminoacylation was moderately affected. When the eukaryal extension was mutated, severe tRNA charging defects were observed, suggesting that eukaryotes evolved this LSD1 extension in order to improve the aminoacylation reaction step. The results also showed that the LSD1s from organisms of both groups are dispensable for post-transfer editing. Together, the data provide us with a further understanding of the organization and structure of LeuRS domains.

INTRODUCTION

The genetic code is dependent on the accurate aminoacylation of tRNAs by aaRSs (aminoacyl-tRNA synthetases) which arose more than 3.5 billion years ago [1,2]. The aminoacylation, which matches the nucleotide triplets embedded in a tRNA molecule (anticodon) with a specific and cognate amino acid, is catalysed in two separate steps by the aaRS [3]. The amino acid is first activated with ATP to produce aminoacyl-adenylate, and the activated amino acid is then transferred to the 3′-end of the tRNA. Occasionally, non-cognate amino acids, whose structures are similar to the cognate one, can be mis-activated and mis-transferred to the tRNAs by some aaRSs [4]. To overcome this problem, some aaRSs evolved proofreading (editing) mechanisms to hydrolyse the incorrect products at the aminoacyl-adenylate level (pre-transfer editing) and/or mis-charged tRNA level (post-transfer editing) [57].

A total of 20 aaRSs comprise two distinct families of enzymes, each of which originated from an ancient distinct single-domain protein [2,8,9]. This domain was believed to contain the active site for amino acid activation and for the subsequent attachment of the activated amino acid to the 3′-end of tRNA [2,8]. The single-domain protein evolved to the extant aaRS through domain fusions at the N- or C-terminus, or domain insertion for higher levels of tRNA binding, aminoacylation efficiency or fidelity, and interaction with other proteins [1013]. A prominent example is LeuRS (leucyl-tRNA synthetase) of the class Ia aaRS [1113]. Other than the ancient Rossmann fold domain common in class I aaRSs, there are several other fused and inserted domains, such as CP (connective peptide) 1 [7] and 2 [9,14], the C-terminal domain and the α-helix bundle domain [9]. In class Ia LeuRS, IleRS (isoleucyl-tRNA synthetase) and ValRS (valyl-tRNA synthetase), the CP1 domain is the editing domain and is responsible for the hydrolysis of the incorrectly synthesized products [57]. The CP2 domain is indispensable for the amino acid activation and post-transfer editing in LeuRSs [14], whereas the C-terminal domain and α-helix bundle domain are essential for the tRNA binding throughout the tRNALeu-charging reaction [12,13,15].

On the basis of the primary sequence, LeuRSs can be divided into two groups: prokaryotic LeuRS and eukaryal/archaeal LeuRS [9]. In the former group, the CP1 domain is located after the CP2 domain, in obvious contrast with the CP1 domain in eukaryal/archaeal LeuRS which precedes the CP2 domain [14]. The editing domains in the two groups are in opposite orientations [9], and the LeuRSs have other domain differences between the two groups [12,13]. For example, a discrete folded LSD (leucine-specific domain) in prokaryotic LeuRS is located just N-terminal to the conserved class I signature sequence KMSKS and affects the aminoacylation turnover number [16]. This unique insert is also the split site for the heterodimeric LeuRS from Aquifex aeolicus [15,17,18]. Similarly, two other LSDs (LSD1 and LSD2) are found in eukaryal and archaeal LeuRSs, but are absent from all prokaryotic LeuRSs [9]. According to the crystal structure of PhLeuRS (Pyrococcus horikoshii LeuRS) (PDB code 1WKB), LSD1 spans from residue 96 to residue 121 and includes two antiparallel α-helices inserted between two α-helices of the Rossmann fold (Figure 1A) [9]. In PhLeuRS in complex with tRNALeu in the aminoacylation conformation (PDB code 1WZ2) (Figure 1B) [12], the 3′-end of tRNA binds the aminoacylation active site without any direct interaction with the LSD1 peptide. In the crystal structure of the apo enzyme, PhLeuRS shows a clear contact between LSD1 and the KMSKS loop, suggesting that LSD1 may have the potential to regulate the KMSKS loop which is highly dynamic and flexible during substrate binding and catalysis [19].

Three-dimensional view of the LeuRS structure and sequence alignment of different eukaryal and archaeal LSD1s

Figure 1
Three-dimensional view of the LeuRS structure and sequence alignment of different eukaryal and archaeal LSD1s

(A) Global view of the PhLeuRS structure (PDB code 1WKB) [9]. The active site containing the Rossmann fold domain and the other accessory domains is coloured green. The editing site appears in yellow and the helix bundle domain is coloured magenta. The LSD1 extension is represented in red. (B) View in the same orientation of the PhLeuRS structure in complex with tRNALeu (PDB code 1WZ2) [12]. The tRNA molecule is coloured orange and the C-terminal domain, visible in the structure, is coloured cyan. The KMSKS loop located near the LSD1 is coloured purple. (C) Sequence alignment of the LSD1s from eukaryotes and archaea. Residues strictly or highly conserved are shadowed. The amino acid residues that have been mutated in the present study are numbered vertically. The sequence of T. thermophilus (Tt) is shown at the bottom of the alignment. It highlights the absence of LSD1 from bacterial LeuRSs. Ce, Caenorhabditis elegans; Hs, Homo sapiens; Mj, Methanococcus jannaschii; Mt, Methanobacterium thermoautotrophicus; Pa, Pyrococuss abyssi; Sc, Saccharomyces cerevisiae. (D) Partial view of the active site showing the LSD1 environment in the active site of PhLeuRS (PDB code 1WKB) [9]. The HIGH loop is coloured blue, the KMSKS loop is magenta and the LSD1 is red. The proximity of Ser46, His48, Lys100, Tyr105, Tyr109 and Ser654 can be seen, highlighting the potential interactions between the LSD1 and the HIGH and KMSKS loops. (E) The active site was rotated by approx. 180 °C. The view highlights the role of Phe119 in the formation of a hydrophobic core that connects LSD1 and the active-site domain. Near Phe119 are found Val111, Ile115 and Thr118 from LSD1 and Phe129 from the adjacent helix of the active site.

Figure 1
Three-dimensional view of the LeuRS structure and sequence alignment of different eukaryal and archaeal LSD1s

(A) Global view of the PhLeuRS structure (PDB code 1WKB) [9]. The active site containing the Rossmann fold domain and the other accessory domains is coloured green. The editing site appears in yellow and the helix bundle domain is coloured magenta. The LSD1 extension is represented in red. (B) View in the same orientation of the PhLeuRS structure in complex with tRNALeu (PDB code 1WZ2) [12]. The tRNA molecule is coloured orange and the C-terminal domain, visible in the structure, is coloured cyan. The KMSKS loop located near the LSD1 is coloured purple. (C) Sequence alignment of the LSD1s from eukaryotes and archaea. Residues strictly or highly conserved are shadowed. The amino acid residues that have been mutated in the present study are numbered vertically. The sequence of T. thermophilus (Tt) is shown at the bottom of the alignment. It highlights the absence of LSD1 from bacterial LeuRSs. Ce, Caenorhabditis elegans; Hs, Homo sapiens; Mj, Methanococcus jannaschii; Mt, Methanobacterium thermoautotrophicus; Pa, Pyrococuss abyssi; Sc, Saccharomyces cerevisiae. (D) Partial view of the active site showing the LSD1 environment in the active site of PhLeuRS (PDB code 1WKB) [9]. The HIGH loop is coloured blue, the KMSKS loop is magenta and the LSD1 is red. The proximity of Ser46, His48, Lys100, Tyr105, Tyr109 and Ser654 can be seen, highlighting the potential interactions between the LSD1 and the HIGH and KMSKS loops. (E) The active site was rotated by approx. 180 °C. The view highlights the role of Phe119 in the formation of a hydrophobic core that connects LSD1 and the active-site domain. Near Phe119 are found Val111, Ile115 and Thr118 from LSD1 and Phe129 from the adjacent helix of the active site.

In the present study, we set out to understand the evolution of LSD1 in eukaryal and archaeal LeuRSs. By sequence alignment, we found that the LSD1s from eukaryotes have evolved additional sequence elements that increased the size of the domain from approx. 26 residues for an archaeal LeuRS to approx. 71 residues for a eukaryal LeuRS. To understand this variation between the kingdoms, the function of the LSD1s from the eukaryote Giardia lamblia and the archaeon P. horikoshii (Gl-LSD1 and Ph-LSD1 respectively) were analysed by sequence deletions and point mutations. The data indicated that both kingdoms share a common region roughly corresponding to the archaeal LSD1 which is involved in amino acid activation and active-site structuring. The eukaryotic-specific extension is involved in tRNA charging, whereas neither the archaeal nor the eukaryal LSD1 is involved in the tRNA-editing process. Altogether, these data support the hypothesis that the LSD1s have a common origin and an intrinsic plasticity for evolving new properties.

EXPERIMENTAL

Materials

L-Leucine, DTT (dithiothreitol), NTP, 5′-GMP, tetrasodium pyrophosphate, PPi, ATP, Tris/HCl, MgCl2, NaCl and activated charcoal were purchased from Sigma. [3H]L-leucine and tetrasodium [32P]pyrophosphate were obtained from PerkinElmer. [3H]L-isoleucine was purchased from GE Healthcare. GF/C filters were purchased from Whatman. Pfu DNA polymerase was obtained from Biotech Company. The DNA fragment rapid purification kit and plasmid extraction kit were purchased from Tiangen Company. T4 ligase and restriction endonucleases were obtained from MBI Fermentas. The anti-His6 antibody was purchased from Sigma. Ni-NTA (Ni2+-nitrilotriacetate) Superflow was purchased from Qiagen. Pyrobest DNA polymerase and the dNTP mixture were obtained from TaKaRa. Oligonucleotide primers were synthesized by Invitrogen. The KOD plus mutagenesis kit was purchased from TOYOBO. The pET28a(+) and pUC19 vectors were acquired from Novagen. T7 RNA polymerase was purified in our laboratory. Escherichia coli KL231 strain [F, leuS31(ts), thyA6, rpsL120(strR), deoC1] [20] was purchased from the E. coli Genetic Stock Center (Yale University, New Haven, CT, U.S.A.).

Cloning and mutagenesis

The deletion and single-point mutants from GlLeuRS (G. lamblia LeuRS) and PhLeuRS were constructed according to the protocol provided by the KOD plus mutagenesis kit using pET28a(+)-GlleuS and pET28a(+)-PhleuS [14] as templates respectively. The genes encoding the chimaeric LSD1-swapping mutants of GlLeuRS–PhLSD1 and PhLeuRS–GlLSD1 were constructed by methods described previously [14]. The gene encoding the His6-tagged GlLeuRS, its LSD1 deletion mutant and its single-point mutants were cleaved by NcoI and BamHI and then inserted into pTrc99B pre-cleaved by NcoI and BamHI to produce pTrc99B-GlleuS and other mutants.

Protein expression and purification

E. coli BL21-Codon Plus(DE3)-RIL cells (Stratagene) were transformed with the plasmids containing the genes encoding GlLeuRS, PhLeuRS and their mutants to overproduce the wild-type LeuRSs and the mutants respectively. A single colony was used to inoculate 500 ml of 2YT [1.6% (w/v) tryptone/1% (w/v) yeast extract/0.5% (w/v) NaCl] medium at 37 °C. Protein overproduction was induced when the cells reached the mid-exponential phase (D600=0.6), by adding IPTG (isopropyl β-D-thiogalactoside) to a final concentration of 0.2 mM. Induction was performed at 22 °C for 6 h for GlLeuRS or 8 h for PhLeuRS. The cells were collected by centrifugation at 10000 g for 30 min and washed twice with a solution of 10 mM imidazole in buffer A (300 mM NaCl, 10% glycerol, 0.5 mM PMSF and 50 mM NaH2PO4, pH 8.0). The GlLeuRS purification was carried out by affinity chromatography on Ni-NTA Superflow (Qiagen). Wet cells from 500 ml of culture (approx. 2 g) were suspended in 10 ml of 10 mM imidazole in buffer A and sonicated on ice. The lysate was centrifuged at 10000 g for 30 min to remove cell debris and the supernatant was ultracentrifuged at 46000 rev./min for 1 h using an 80 Ti rotor in a Beckman Coulter ultracentrifuge to remove the ribosomes. The supernatant was gently mixed with 1 ml of Ni-NTA Superflow resin for 1 h. Then, the suspension was poured in a column and the resin was washed with 30 ml of a 20 mM imidazole solution in buffer A in order to remove non-specific binding proteins. Then, the enzyme was eluted in two steps by adding 7 ml of 50 mM imidazole solution in buffer A and 8 ml of 250 mM imidazole in buffer A. The fractions containing the eluted proteins were pooled, dialysed against 50 mM Tris/HCl (pH 8.2), 50 mM NaCl and 2 mM 2-mercaptoethanol solution for GlLeuRS and concentrated on an Amicon Ultra-15 filter (Millipore). The PhLeuRS purification procedure was the same as the GlLeuRS procedure except that an additional 10-min heating step at 65 °C was performed on the sonicated crude extracts and the pH of the dialysis buffer that was 7.5 for PhLeuRS was 8.2 for GlLeuRS. All purification steps were carried out at 4 °C. In a last step, the purified enzymes were gently mixed with an equal volume of glycerol and stored at −20 °C.

Preparation of tRNA and mis-charged tRNA

GltRNALeu [G. lamblia tRNALeu(AAG)] and PhtRNALeu [P. horikoshii tRNALeu(GAG)] were prepared by in vitro T7 RNA polymerase transcription as described in [14]. tRNAs from GltRNALeu and PhtRNALeu mis-charged with isoleucine were obtained with the editing-deficient mutants GlLeuRS-D444A [14] and PhLeuRS-D332A [9] by incubation for 30 min at 45 °C and 65 °C respectively. The mis-charged tRNAs were purified by repeated phenol/chloroform extractions, followed by ethanol precipitation.

ATP–PPi exchange and aminoacylation assays

The ATP–PPi exchange reactions catalysed by GlLeuRSs were carried out in reaction mixtures containing 60 mM Tris/HCl (pH 8.2), 10 mM MgCl2, 2 mM DTT, 4 mM ATP, 1 mM leucine, 2 mM tetrasodium [32P]pyrophosphate and 20 nM enzyme at 45 °C. For the determination of the Michaelis constant (Km) of ATP by GlLeuRS, the reaction mixture was essentially the same except that the ATP concentrations varied between 0.2 and 5 mM. For determining the Km of leucine, the concentration of leucine varied between 2.6 and 42.8 μM. The reactions catalysed by PhLeuRS and its mutants were performed in reaction mixtures containing 60 mM Tris/HCl (pH 7.5), 10 mM MgCl2, 2 mM DTT, 4 mM ATP, 2 mM leucine, 2 mM tetrasodium [32P]pyrophosphate and 100 nM of LeuRSs at 65 °C. To determine the Km for ATP by PhLeuRS, the concentration of ATP varied between 0.2 and 2.5 mM. The Km for leucine was determined in the same reaction mixture with various concentrations of leucine between 1 and 40 μM. Aliquots (15 μl) of the reaction mixtures were removed at different time intervals, and the enzymatic reactions were stopped using 200 μl of quenching buffer (50 mM tetrasodium pyrophosphate, 2% activated charcoal and 3.5% HClO4). The charcoal suspension was filtered through a Whatman GF/C filter, washed three times with 5 ml of water and rinsed with 5 ml of 100% ethanol. The charcoal powder on the filters was dried, and the synthesized [32P]ATP was counted using a scintillation counter (Beckman Coulter).

The aminoacylation assay catalysed by GlLeuRS was carried out as described previously [21]. The aminoacylation assay catalysed by PhLeuRSs was performed in a reaction mixture containing 60 mM Tris/HCl (pH 7.5), 10 mM MgCl2, 2 mM DTT, 4 mM ATP, 40 μM [3H]leucine, 10 μM PhtRNALeu and 100 nM PhLeuRS at 37 °C. The kinetic constants for tRNA by GlLeuRS were measured by varying the GltRNALeu concentration between 0.28 and 14 μM. The kinetic constants for tRNA by PhLeuRS were measured by varying the PhtRNALeu concentration between 0.48 and 15.5 μM with 20 nM PhLeuRS at 37 °C. Aliquots (15 μl) of the reaction mixtures were removed at specific time intervals and quenched on Whatman filter pads in 5% trichloroacetic acid solution. The pads were washed three times for 5 min each with 5% trichloroacetic acid solution and then rinsed with 100% ethanol. The pads were dried, and the radioactivity of the precipitates was quantified using a scintillation counter. All of the kinetic constants were calculated by Hanes–Woolf plot (see Supplementary Figure S1 at http://www.BiochemJ.org/bj/429/bj4290505add.htm).

Complementation assay in the thermosensitive E. coli KL231 strain

Competent cells were transformed with various plasmids, including pTrc99B (negative control), pTrc99B-EcleuS, pTrc99B-GlleuS, pTrc99B-GlleuS-ΔLSD1, -K139A, -K141A, -K142A, -K144A and -F171A [22]. For each plasmid, a single colony was selected and grown in liquid LB (Luria–Bertani) medium supplemented with 100 μg/ml ampicillin and 200 μg/ml thymine at 30 °C. The cells were grown overnight, diluted and adjusted to a D600 of 0.2, and then 5 μl of each diluted culture was dropped on to two LB plates supplemented with 100 μg/ml ampicillin and 200 μg/ml thymine. The two plates were incubated at 30 °C and 42 °C respectively, then the cell growth was observed and compared.

Western blot analysis

KL231 cells containing the genes encoding the wild-type and mutated GlLeuRSs were grown at 30 °C overnight. The cells were harvested by centrifugation at 5000 g for 30 min, lysed by sonication, and cell debris was removed by centrifugation at 10000 g for 30 min. Equal amounts of the protein-containing supernatants (60 μg) were separated by SDS/PAGE (7.5% gel) and the gel was electroblotted on to a PVDF membrane (NEN Research Products). The membrane was blocked with 5% (w/v) non-fat dried milk in PBST (1× PBS plus 0.05% Triton X-100) and incubated overnight with a 1:2000 dilution of anti-His6 antibody at 4 °C. Then, a 1:5000 dilution of the secondary goat anti-(mouse IgG) conjugated to horseradish peroxidase was added and incubated at room temperature (25 °C) for 2 h. The bound antibody was visualized by adding the Western Lightning® horseradish peroxidase chemiluminescence solution (PerkinElmer) and exposing to Fujifilm LAS-4000 (Fujifilm Life Science).

RESULTS

Sequence alignments revealed an increase in the size of eukaryal LSD1s

GlLeuRS and PhLeuRS are representative members of the eukaryal and archaeal kingdoms and consist of 1173 and 967 amino acid residues respectively. On the basis of the PhLeuRS three-dimensional structure (PDB code 1WKB) and sequence alignments, we first identified one of the ‘leucine-specific domains’ of GlLeuRS, named Gl-LSD1, which is only found in archaeal and eukaryal enzymes (Figure 1C). Sequence analysis revealed that archaeal and eukaryal LSD1s are markedly different, with those from eukaryal LeuRSs being longer (approx. 71 residues) than those from archaea (26 residues). The eukaryal LSD1 was extended by approx. 45 amino acids in the N-terminal part, and this extension is rich in highly conserved polar amino acid residues, especially several consecutive lysine residues (Figure 1C). On the other hand, eukaryal and archaeal sequences show significant residue conservation in the second half of the LSD1, suggesting that they adopt similar three-dimensional folding. This part of LSD1 can be considered as the common eukaryal/archaeal ancestral domain to which the eukaryal extension was added during evolution. In PhLeuRS, this ancestral domain, or core domain, contains the two antiparallel α-helices that were observed to interact with the KMSKS motif in the crystal structure of the apo enzyme [9].

Deletion of the entire LSD1s abolishes synthetic activity of the two LeuRSs

As a first approach to deciphering the function of the two LSD1s, we performed deletions of Gl-LSD1 (eukaryal LeuRS) and Ph-LSD1 (archaeal LeuRS) in the two corresponding enzymes. The deletions extended from Asp102 to Asp173 in GlLeuRS and Asn96 to Asp121 in Ph-LSD1 (Figure 1C). A four-alanine linker peptide was inserted at the deletion point in PhLeuRS and GlLeuRS in order to allow a proper connection of the secondary-structure elements and favour folding of the mutants.

The two mutated proteins were overexpressed in E. coli and purified as soluble proteins, and the leucine activation was tested by an ATP–PPi exchange reaction. The two mutants were apparently unable to significantly activate leucine (Figure 2). Similarly, tRNA charging was completely defective (Figure 2). Together, these assays suggested that the LSD1s are essential for the synthesis activity. Meanwhile, the post-transfer editing activity was tested using the deacylation assay of the mis-charged Ile-tRNALeu. The editing activity of PhLeuRS-ΔLSD1 was identical with that of the wild-type PhLeuRS, indicating that the global conformation of the enzyme was preserved and that binding of tRNA in the editing conformation was unaffected by the LSD1 deletion (Figure 2). In contrast, for GlLeuRS, the editing activity was nearly abolished (Figure 2). These results suggested that Gl-LSD1 might also be responsible for the post-transfer editing activity or, alternatively, that the conformation of the deletion mutant was altered.

Enzymatic activities analysis of wild-type and LSD1-deleted LeuRS from G. lamblia (A–C) and P. horikoshii (D–F)

Figure 2
Enzymatic activities analysis of wild-type and LSD1-deleted LeuRS from G. lamblia (A–C) and P. horikoshii (D–F)

Wild-type LeuRS curves (●) are compared with LSD1-deleted curves (○). (A and D) ATP formation in the ATP–PPi exchange assay. (B and E) Leu-tRNALeu formation. (C and F) Deacylation activity of mis-charged Ile-tRNALeu. As a control, non-enzymatic hydrolysis of Ile-tRNALeu (spontaneous hydrolysis) is also shown (▼).

Figure 2
Enzymatic activities analysis of wild-type and LSD1-deleted LeuRS from G. lamblia (A–C) and P. horikoshii (D–F)

Wild-type LeuRS curves (●) are compared with LSD1-deleted curves (○). (A and D) ATP formation in the ATP–PPi exchange assay. (B and E) Leu-tRNALeu formation. (C and F) Deacylation activity of mis-charged Ile-tRNALeu. As a control, non-enzymatic hydrolysis of Ile-tRNALeu (spontaneous hydrolysis) is also shown (▼).

Alanine scanning of the eukaryotic-specific LSD1 extension

To investigate the function of the conserved positively charged residues in the eukaryal LSD1 extensions, five lysine residues were mutated in GlLeuRS in order to give the mutants GlLeuRS-K139A, -K141A, -K142A, -K144A and -K148A. The mutated proteins were overexpressed in E. coli and purified to near homogeneity. Then, steady-state kinetic parameters were measured and in vivo complementation assays were performed to establish a correlation between the in vivo and in vitro effects.

A significant decrease in the tRNA-charging activity was observed with the first four mutants; however, the last one, K148A, showed no significant change. The catalytic efficiencies of the four mutants were reduced from 3- to 11-fold when compared with the wild-type LeuRS (Table 1). These effects were due to decreases in the kcat and in one case to a simultaneous increase in the Km for tRNA (K139A). The amino acid activation activities of the five mutants were unchanged as well as the post-transfer editing capacity measured by the hydrolysis of preformed Ile-tRNALeu (Table 2 and results not shown). In addition, these mutants were unable to mis-charge tRNALeu with the non-cognate isoleucine, which confirmed that the editing properties of the mutants were intact (results not shown).

Table 1
Steady-state aminoacylation kinetics of various mutants from GlLeuRS

Each assay was carried out three times under the same conditions. Data for GlLeuRS are from [21]. NM, not measurable. Significant differences in kinetic constants from GlLeuRS are indicated by *.

Enzyme Km (μM) kcat (s−1kcat/Km (s−1·mM−1Relative kcat/Km 
GlLeuRS 6.1±0.4 2.7±0.2 443 
K139A 11.4±2.1* 1.5±0.1* 132* 0.30* 
K141A 5.1±0.6 0.8±0.1* 157* 0.35* 
K142A 3.5±0.5 0.2±0.03* 57* 0.13* 
K144A 5.2±0.7 0.2±0.04* 38* 0.09* 
K148A 5.0±0.6 1.9±0.3 380 0.86 
K152A 5.1±0.4 1.3±0.2* 255* 0.58 
S153A 4.0±0.4 1.1±0.2* 275* 0.62 
Q154A 3.8±0.5 0.8±0.1* 210* 0.48 
W155A 7.1±0.9 2.5±0.4 352 0.79 
E165A 8.3±0.9 2.8±0.5 337 0.76 
K166A 8.1±1.0 2.8±0.5 346 0.78 
E167A 4.9±0.6 1.2±0.3* 245* 0.55 
K170A 5.1±0.7 1.4±0.3* 275* 0.62 
F171A NM NM NM NM 
D173A 5.9±0.8 0.8±0.2* 136* 0.31 
Enzyme Km (μM) kcat (s−1kcat/Km (s−1·mM−1Relative kcat/Km 
GlLeuRS 6.1±0.4 2.7±0.2 443 
K139A 11.4±2.1* 1.5±0.1* 132* 0.30* 
K141A 5.1±0.6 0.8±0.1* 157* 0.35* 
K142A 3.5±0.5 0.2±0.03* 57* 0.13* 
K144A 5.2±0.7 0.2±0.04* 38* 0.09* 
K148A 5.0±0.6 1.9±0.3 380 0.86 
K152A 5.1±0.4 1.3±0.2* 255* 0.58 
S153A 4.0±0.4 1.1±0.2* 275* 0.62 
Q154A 3.8±0.5 0.8±0.1* 210* 0.48 
W155A 7.1±0.9 2.5±0.4 352 0.79 
E165A 8.3±0.9 2.8±0.5 337 0.76 
K166A 8.1±1.0 2.8±0.5 346 0.78 
E167A 4.9±0.6 1.2±0.3* 245* 0.55 
K170A 5.1±0.7 1.4±0.3* 275* 0.62 
F171A NM NM NM NM 
D173A 5.9±0.8 0.8±0.2* 136* 0.31 
Table 2
Steady-state leucine activation kinetics of GlLeuRS and its mutants

Each assay was carried out three times under the same conditions. Data for GlLeuRS are from [21]. NM, not measurable. Significant differences in kinetic constants from GlLeuRS are indicated by *.

 ATP Leucine 
Enzyme Km (μM) kcat (s−1kcat/Km (s−1·mM−1Km (μM) kcat (s−1kcat/Km (s−1·mM−1
GlLeuRS 750±61 34.5±2.3 46.0 14.2±1.5 30.5±3.5 2148 
K139A 812±95 33.2±3.8 40.9 15.1±2.4 29.9±4.1 1980 
K141A 772±66 31.5±2.9 40.8 13.2±1.9 33.1±2.8 2508 
K142A 728±87 36.2±3.1 49.7 14.5±2.7 31.7±2.5 2186 
K144A 822±91 31.8±3.6 38.7 15.3±3.0 28.9±3.5 1889 
K148A 698±87 32.1±4.0 46.0 12.4±2.4 28.4±3.8 2290 
K152A 834±86 8.1±0.9* 9.7 2.4±0.3* 6.2±0.8* 2583 
S153A 711±89 30.1±2.9 42.3 15.9±4.1 32.5±2.1 2044 
Q154A 675±65 7.5±0.6* 11.1 3.5±0.5* 5.9±0.8* 1686 
W155A 653±71 29.8±3.7 45.6 11.9±2.8 31.9±3.3 2681 
E165A 1169±130 31.3±4.2 26.8 15.7±2.3 29.7±3.7 1892 
K166A 687±74 32.3±3.5 47.0 14.7±3.7 28.7±3.0 1952 
E167A 2129±197* 60.2±7.7* 28.3 16.2±3.1 54.6±6.9 3370 
K170A 683±78 31.0±3.4 45.4 14.1±2.8 28.1±2.3 1993 
F171A NM NM NM NM NM NM 
D173A 1025±94 34.4±4.2 33.5 16.5±3.1 36.3±4.6 2200 
 ATP Leucine 
Enzyme Km (μM) kcat (s−1kcat/Km (s−1·mM−1Km (μM) kcat (s−1kcat/Km (s−1·mM−1
GlLeuRS 750±61 34.5±2.3 46.0 14.2±1.5 30.5±3.5 2148 
K139A 812±95 33.2±3.8 40.9 15.1±2.4 29.9±4.1 1980 
K141A 772±66 31.5±2.9 40.8 13.2±1.9 33.1±2.8 2508 
K142A 728±87 36.2±3.1 49.7 14.5±2.7 31.7±2.5 2186 
K144A 822±91 31.8±3.6 38.7 15.3±3.0 28.9±3.5 1889 
K148A 698±87 32.1±4.0 46.0 12.4±2.4 28.4±3.8 2290 
K152A 834±86 8.1±0.9* 9.7 2.4±0.3* 6.2±0.8* 2583 
S153A 711±89 30.1±2.9 42.3 15.9±4.1 32.5±2.1 2044 
Q154A 675±65 7.5±0.6* 11.1 3.5±0.5* 5.9±0.8* 1686 
W155A 653±71 29.8±3.7 45.6 11.9±2.8 31.9±3.3 2681 
E165A 1169±130 31.3±4.2 26.8 15.7±2.3 29.7±3.7 1892 
K166A 687±74 32.3±3.5 47.0 14.7±3.7 28.7±3.0 1952 
E167A 2129±197* 60.2±7.7* 28.3 16.2±3.1 54.6±6.9 3370 
K170A 683±78 31.0±3.4 45.4 14.1±2.8 28.1±2.3 1993 
F171A NM NM NM NM NM NM 
D173A 1025±94 34.4±4.2 33.5 16.5±3.1 36.3±4.6 2200 

We showed recently that GlLeuRS catalyses the E. coli tRNALeu and GltRNALeu esterification with the same efficiency [21]. To explore the significance of this phenomenon in vivo and to confirm it using a functional assay, we used the E. coli strain KL231 carrying a temperature-sensitive LeuRS [20] in a complementation assay with GlLeuRS proteins. We cloned the genes encoding GlLeuRS and single-point mutants K139A, K141A, K142A and K144A into pTrc99B. The KL231 strain was transformed with the recombinant plasmids, and growth was compared at permissive and non-permissive temperatures. At 30 °C, the wild-type GlLeuRS and its mutants were expressed to similar levels (Figure 3A) and all of the transformants grew on medium supplemented with 100 μg/ml ampicillin and 200 μg/ml thymine (Figure 3B). However, at 42 °C, when the endogenous LeuRS was inactivated, substantial differences were observed. Whereas the cell growth was normal with wild-type GlLeuRS, it was impaired with the mutants K139A, K141A, K142A and K144A (Figure 3B). The growth defect was more obvious with the two last mutants which showed the lowest activity of tRNA charging in vitro (9 and 13% respectively; Table 1). The loss of charging activity was likely to be responsible for the absence of complementation of the thermosensitive E. coli strain. Therefore the in vivo data are consistent with the in vitro data and confirmed the functional importance of the four lysine residues, and especially of the lysine residues at positions 142 and 144.

Complementation assay of the E. coli KL231 strain

Figure 3
Complementation assay of the E. coli KL231 strain

(A) Western blot analysis on E. coli KL231 strains expressing GlLeuRS proteins at 30 °C. All of the LeuRSs proteins were expressed equally. (B) Genes encoding different GlLeuRS mutants, wild-type E. coli LeuRS, wild-type GlLeuRS (as positive controls) and pTrc99B (as negative control) were introduced into E. coli leuS temperature-sensitive strain KL231. Strains expressing the different LeuRS genes were grown at 30 and 42 °C in order to test the in vivo capacity of the mutants to complement the bacterial LeuRS activity.

Figure 3
Complementation assay of the E. coli KL231 strain

(A) Western blot analysis on E. coli KL231 strains expressing GlLeuRS proteins at 30 °C. All of the LeuRSs proteins were expressed equally. (B) Genes encoding different GlLeuRS mutants, wild-type E. coli LeuRS, wild-type GlLeuRS (as positive controls) and pTrc99B (as negative control) were introduced into E. coli leuS temperature-sensitive strain KL231. Strains expressing the different LeuRS genes were grown at 30 and 42 °C in order to test the in vivo capacity of the mutants to complement the bacterial LeuRS activity.

Analysis of the ancestral core of LSD1

Crystal structures of both apo-PhLeuRS (PDB code 1WKB) [9] and PhLeuRS-tRNALeu aminoacylation complex (PDB code 1WZ2) [12] revealed that Ph-LSD1 is a small domain containing two antiparallel α-helices. This domain is within interacting distance from the KMSKS loop, but too far from the acceptor branch of the tRNA (Figure 1B). It was suggested that Ph-LSD1 might provide a docking site for the flexible KMSKS loop and thereby facilitate its dynamics [9]. The sequence alignments shown in Figure 1(C) strongly suggest that these antiparallel α-helices are also conserved in eukaryotic LeuRSs along with the new eukaryotic-specific extension. To assay the importance of the residues from the ancestral part of LSD1, 17 out of the 26 residues present in archaeal PhLeuRS were mutated. Ten residues were also mutated in the eukaryal GlLeuRS. The mutations constructed in PhLeuRS were N96A, R97A, D98A, K100A, T101A, W103A, I104A, Y105A, R106A, V108A, Y109A, E113A, E114A, I115A, T118A, F119A and D121A. In GlLeuRS, the mutations were K152A, S153A, Q154A, W155A, E165A, K166A, E167A, K170A, F171A and D173A. Eight of these mutations corresponded to residues also mutated in PhLeuRS (Figure 1C).

In the leucine activation reaction, several mutations in PhLeuRS induced variations in the kinetic parameters as measured in the ATP–PPi exchange assay (Table 3). PhI115A and PhT118A exhibited 4- and 2-fold increases in the Km for ATP respectively. Remarkably, the increase in Km for ATP of PhI115A was followed by an increase in the kcat. PhR106A displayed a 2-fold decrease in the Km for ATP and PhF119A showed a 2-fold increase in the Km for leucine, leading to an enzyme with the weakest catalytic efficiency.

Table 3
Steady-state leucine activation kinetics of PhLeuRS and its single-point mutants

Each assay was carried out three times under the same conditions. Significant differences in kinetic constants from GlLeuRS are indicated by *.

 ATP Leucine 
Enzyme Km (μM) kcat (s−1kcat/Km (s−1·mM−1Km (μM) kcat (s−1kcat/Km (s−1·mM−1
PhLeuRS 578±72 6.8±0.8 11.8 5.92±0.8 6.0±0.5 1013 
N96A 370±54 5.1±0.6 13.8 5.37±0.6 4.4±0.4 819 
R97A 323±52 5.5±0.7 17.0 9.27±1.3 5.1±0.6 550 
D98A 362±49 4.3±0.5 11.9 5.92±0.6 3.2±0.4 541 
K100A 531±78 4.6±0.6 8.7 8.91±0.7 3.9±0.5 438 
T101A 383±45 5.0±0.7 13.1 9.84±0.8 3.8±0.7 386 
W103A 472±62 7.4±0.9 15.7 8.71±0.5 5.1±0.6 586 
I104A 796±89 7.4±0.7 9.3 5.69±0.8 5.6±0.5 984 
Y105A 537±65 6.5±0.8 12.1 5.55±0.7 4.3±0.5 775 
R106A 296±34* 5.4±0.4 18.2 8.54±0.5 4.7±0.4 550 
V108A 351±53 4.2±0.6 12.0 5.81±0.3 4.1±0.3 706 
Y109A 591±66 4.8±0.5 8.1 9.02±0.8 4.7±0.5 521 
E113A 837±74 5.8±0.7 6.9 6.18±0.9 5.2±0.7 846 
E114A 373±38 4.3±0.7 11.5 6.23±0.8 3.1±0.2 498 
I115A 2177±184* 13.8±2.2* 6.4 5.59±0.6 10.7±1.9 1910 
T118A 1157±126* 7.7±1.4 6.6 6.36±0.7 8.1±1.2 1277 
F119A 547±60 4.0±0.6 7.3 11.8±1.4* 3.4±0.5 286* 
D121A 551±67 4.7±0.8 8.5 7.21±0.9 6.0±1.1 839 
K100A/Y105A 584±62 3.0±0.5 5.2 6.3±0.6 3.1±0.4 493 
K100A/Y109A 366±54 4.9±0.7 13.5 5.7±0.7 4.6±0.6 806 
 ATP Leucine 
Enzyme Km (μM) kcat (s−1kcat/Km (s−1·mM−1Km (μM) kcat (s−1kcat/Km (s−1·mM−1
PhLeuRS 578±72 6.8±0.8 11.8 5.92±0.8 6.0±0.5 1013 
N96A 370±54 5.1±0.6 13.8 5.37±0.6 4.4±0.4 819 
R97A 323±52 5.5±0.7 17.0 9.27±1.3 5.1±0.6 550 
D98A 362±49 4.3±0.5 11.9 5.92±0.6 3.2±0.4 541 
K100A 531±78 4.6±0.6 8.7 8.91±0.7 3.9±0.5 438 
T101A 383±45 5.0±0.7 13.1 9.84±0.8 3.8±0.7 386 
W103A 472±62 7.4±0.9 15.7 8.71±0.5 5.1±0.6 586 
I104A 796±89 7.4±0.7 9.3 5.69±0.8 5.6±0.5 984 
Y105A 537±65 6.5±0.8 12.1 5.55±0.7 4.3±0.5 775 
R106A 296±34* 5.4±0.4 18.2 8.54±0.5 4.7±0.4 550 
V108A 351±53 4.2±0.6 12.0 5.81±0.3 4.1±0.3 706 
Y109A 591±66 4.8±0.5 8.1 9.02±0.8 4.7±0.5 521 
E113A 837±74 5.8±0.7 6.9 6.18±0.9 5.2±0.7 846 
E114A 373±38 4.3±0.7 11.5 6.23±0.8 3.1±0.2 498 
I115A 2177±184* 13.8±2.2* 6.4 5.59±0.6 10.7±1.9 1910 
T118A 1157±126* 7.7±1.4 6.6 6.36±0.7 8.1±1.2 1277 
F119A 547±60 4.0±0.6 7.3 11.8±1.4* 3.4±0.5 286* 
D121A 551±67 4.7±0.8 8.5 7.21±0.9 6.0±1.1 839 
K100A/Y105A 584±62 3.0±0.5 5.2 6.3±0.6 3.1±0.4 493 
K100A/Y109A 366±54 4.9±0.7 13.5 5.7±0.7 4.6±0.6 806 

In GlLeuRS, four mutations (K152A, Q154A, E167A and F171A) showed defects in leucine activation (Table 2). A surprising increase in the leucine affinity was measured with GlK152A and GlQ154A (6- and 4-fold decreases in Km for leucine respectively), as well as a 5-fold decrease in the amino acid activation rate. These data suggest that tightening the interactions with leucine may have prevented optimal stabilization of the transition state of the reaction as proposed previously [23]. Similarly, mutation GlE167A induced a 3-fold increase in the Km for ATP together with an increase in the kcat. The last mutant, GlF171A, was totally inactive in the leucine activation reaction.

Among all of these mutants, only PhF119A and GlF171A affected the equivalent residue in both proteins (Figure 1C). All of the other mutations affected unilaterally the activity of one single enzyme without affecting significantly the other protein. This fact suggested that, although some sequence conservation exists between the ancestral part of the LSD1s of eukaryotes and archaea, the conserved residues did not necessarily always conserve their original function in the two kingdoms.

A different picture emerged from the analysis of the tRNA-charging activity (Table 1 and Supplementary Table S1 at http://www.BiochemJ.org/bj/429/bj4290505add.htm). All PhLeuRS mutants exhibited wild-type aminoacylation activities (Supplementary Table S1), despite the variations of leucine activation activity observed for some of them (see above). This suggested that the rate of the first step of amino acid activation was not limiting and was higher than the rate of the second step of tRNA charging. For eukaryotic GlLeuRS, five mutants (K152A, S153A, Q154A, E167A and D173A) showed decreases in the kcat of the tRNA-charging reaction of approx. 2–3-fold (Table 1) and F171A was totally inactive in vitro (Table 1) and in vivo (Figure 3B).

Overall, the mutational analysis suggested that the ancestral part of LSD1 corresponding to the archaeal LSD1 is involved in the substrate binding and/or catalysis in amino acid activation step. In both PhLeuRS and GlLeuRS, mutants could be identified that had an impact on the activation step. In addition, some mutations reduced the tRNA-charging ability of GlLeuRS, whereas the tRNA aminoacylation ability of PhLeuRS was relatively insensitive to mutations in LSD1.

LSD1 residues do not contribute to stabilization of the KMSKS conformation

In the 1WKB crystal structure of PhLeuRS, residues Lys653 and Ser654 from the K650MSKS654 loop occupy strategic positions for interaction with residues from the first helix of the LSD1. The hydroxy group of Ser654 is located at a distance of only 3.90 Å (1 Å=0.1 nm) from the ε-NH2 group of Lys100, and Lys653 is at distances of 4.69 and 3.69 Å from Tyr105 and Tyr109 of the LSD1 respectively (Figure 1D). It was proposed that the LSD1 residues might interact with the K650MSKS654 loop in one of its highly dynamic conformations [9]. Similarly, the hydroxy group of Tyr105 is located at a distance of only 2.61Å from the main chain from Ser46 located in the loop containing the H48IGH51 sequence, and Tyr109 is located at less than 3.5 Å from His48 (Figure 1D). Thus these LSD1 residues are ideally located to interact with the two crucial loops of the active site. In the present study, Lys100, Tyr105 and Tyr109 were mutated to alanine (see above); however, the activities of leucine activation, aminoacylation and post-transfer editing of these single-point mutants were indistinguishable from those of wild-type PhLeuRS, suggesting that mutation of the individual residues had no significant effect on the enzymatic functions (Table 3 and Supplementary Table S1). Consequently, two double-point mutants, PhK100A/Y105A and PhK100A/Y109A, were constructed in order to visualize the effects of coupling the single mutations. The kcat for leucine activation activity was reduced 2-fold with the PhK100A/Y105A mutant, whereas the PhK100A/Y109A mutant exhibited 75% of the wild-type PhLeuRS activity (Table 3). These effects were modest compared with those observed with some single-point mutations, such as E114A and F119A, which are not in the immediate environment of the KMSKS loop, suggested that residues 100, 105 and 109 do not establish critical interactions with residues from this loop.

Archaeal and eukaryal LSD1 exchanges

In order to determine whether the small archaeal LSD1 could functionally replace the eukaryal LSD1 in GlLeuRS, the amino acids from Asn96 to Asp121 of Ph-LSD1 were inserted between amino acids Ala101 to Pro174 of GlLeuRS (Figure 1C). A comparable substitution of the archaeal LSD1 by the eukaryal LSD1 was performed in PhLeuRS. The two chimaeric proteins were efficiently produced in E. coli and purified. The thermostabilities of the chimaeras were assayed at 50 °C for GlLeuRS and 70 °C for PhLeuRS (results not shown). The thermal denaturation curves were nearly identical when comparing the chimaeric and the corresponding wild-type proteins, confirming that the global enzyme conformation was unchanged. Nevertheless, the adenylate formation activity of the two mutants was completely lost (Figure 4A). On the other hand, they could perform the post-transfer editing function with the same catalytic efficiency as compared with their wild-type LeuRS (Figure 4B). These results showed that, although the LSD1 exchanges retained intact post-transfer editing functions and the thermal resistance of the mutated enzymes, the synthesis activities were completely lost. Domain exchange between proteins is a useful way to investigate their function. However, interpreting the present loss of activity is constrained by the fact that the domain exchanges can modulate the three-dimensional structure of the protein and thus the enzyme activity by indirect effects. The findings of the present study suggest that the LSD1s do not play a critical role in editing, but concluding that the domain plays a crucial role in adenylate formation will require additional structural investigation.

Enzymatic activities of two LSD1 chimaeric enzymes

Figure 4
Enzymatic activities of two LSD1 chimaeric enzymes

(A) ATP–PPi exchange activity and (B) post-transfer editing activity of GlLeuRS (●), GlLeuRS–PhLSD1 (○), PhLeuRS (▼) and PhLeuRS–GlLSD1 (Δ). The spontaneous hydrolysis of Ile-GltRNALeu (■) and Ile-PhtRNALeu is shown (□).

Figure 4
Enzymatic activities of two LSD1 chimaeric enzymes

(A) ATP–PPi exchange activity and (B) post-transfer editing activity of GlLeuRS (●), GlLeuRS–PhLSD1 (○), PhLeuRS (▼) and PhLeuRS–GlLSD1 (Δ). The spontaneous hydrolysis of Ile-GltRNALeu (■) and Ile-PhtRNALeu is shown (□).

DISCUSSION

The small LSD from archaea

LeuRS is closely related to IleRS and ValRS, and together they form the class I LIV-RS subgroup characterized by an active site which is folded according to the Rossmann model on which two additional domains are invariably appended. One of these domains, forming a large helix bundle domain, is involved in tRNA binding. The second one, or CP1 domain, forms a discrete domain involved in amino acid editing [13]. Although the CP1 domain is invariably found, its insertion point between the secondary structure elements of the Rossmann fold may vary, leading to variations in the orientation of the different CP1s relative to the active site [9]. Besides these major additional domains, peripheral insertions vary considerably, depending on the enzyme and the phylogenetic classification of each LIV-RS enzyme. Up to five additional domains have been described in LeuRS (CP-core, CP2, LSD, SC-fold and C-terminal domain) [9]. In archaeal LeuRS, two LSDs, varying in size and insertion points have been described (LSD1 and LSD2) [9]. Of these two domains, LSD2 is located on the opposite side of the enzyme relative to the aminoacylation active site and is unlikely to be able to play a role in the aminoacylation reaction [9]. On the other hand, LSD1 was proposed to have a potential role in contacting tRNA or regulating the aminoacylation reaction because of its close proximity to the KMSKS loop [9]. This element is one of the most flexible loops during amino acid activation, tRNA 3′-end entry and aminoacyl transfer step [24]. The successive motions of the KMSKS loop and the proximity of LSD1 of PhLeuRS prompted us to analyse the hypothetical function of the three residues located in the first helix of the LSD1 that are likely to interact with the KMSKS loop (Lys100, Tyr105 and Tyr109). However, the corresponding alanine mutants did not show significant effects and coupling the mutations had only a moderate impact on the amino acid activation rate. LSD1 adopts a helix–loop–helix structure that partially masks the part of the active site containing the two consensus sequences HIGH and KMSKS. At this position, LSD1 may regulate the enzyme activity by indirect effects. It may stabilize the conformation of the two long helices of the active site between which LSD1 is inserted. Analysis of the three-dimensional structure showed that several residues from the second helix of LSD1 might play crucial roles. For instance, Phe119 which showed the strongest effects of alanine mutation (in PhLeuRS and GlLeuRS) is at the centre of a hydrophobic core formed by residues from LSD1 and the next long helix from the active site (Figure 1E). Similarly, Ile115 is located in the same hydrophobic environment. When mutated to alanine, the enzyme exhibited a loss of affinity for ATP together with an increase in the catalytic constant. Likewise, Thr118 might reinforce the connection of the second helix of LSD1 with the next helix of the active site by interacting with Asn123. Mutant T118A induced similar effects as I115A with a loss of ATP affinity and a slight increase in the catalytic constant. Thus one putative function of LSD1 would be to stabilize the active-site structure, which could explain that Ile115, Thr118 and Phe119, although located far from the KMSKS loop, are able to affect leucine activation.

The large LSD from eukaryotes

Aminoacyl-tRNA synthetases are well known for their function in activating amino acids for protein synthesis. In addition, there are more and more examples of expansion of their activities in various pathways [25,26] by adding new domains, reprogramming existing domains [26] or expropriating existing domains [16]. This wide range of activities may explain why many tRNA synthetase mutations are associated with diseases, sometimes with no change of the aminoacylation activity [2729]. One way that new functions can be developed is through acquisition of appended domains. During evolution, peptide insertions occur in the active site of LeuRSs in order to evolve the LSD1s. Absent from bacteria, LSD1 evolved in eukaryotes and archaea into two different types of domains that share a common C-terminal region and a specific eukaryal appended N-terminal domain for which no structural information is yet available. Our results revealed that the eukaryotic appended domain from G. lamblia contains a set of four proximal lysine residues (Lys139, Lys141, Lys142 and Lys144) that are involved in the aminoacylation of tRNA. The alanine substitutions induced losses of tRNA-charging ability of the four mutants. However, three of them (Lys139, Lys141 and Lys142) were tolerant to arginine, but not to glutamate substitution (results not shown), suggesting some plasticity in their role. The last lysine residue (Lys144) could not be replaced by either arginine or glutamate (results not shown) and thus should occupy a more crucial position in the eukaryal LSD1. At the same time, the alanine mutations did not induce any change in the amino acid activation activity (Table 2), which is particularly sensitive to motions of the KMSKS loop [24]. This point suggests that these four lysine residues do not contribute to the dynamics of the KMSKS loop, but are more likely to contribute to the second step of transfer of the amino acid on tRNALeu. PhLeuRS and GlLeuRS share most of the residues involved in Ade73 and Cyt75 binding of tRNALeu. In the active site of PhLeuRS, the critical residues are Asp504, Lys505, Ala507, Glu527, His81, Ser528, Leu529 and Asn616 [9]. In GlLeuRS, identical residues are found at Ala507, Glu527, His81, Ser528, Leu529 and Asn616 (alignment not shown), suggesting that the tRNA-binding mode is mainly conserved between the eukaryal and archaeal LeuRSs. Consequently, the four lysine residues at positions 139, 141, 142 and 144 should participate in a eukaryotic-specific network of interactions that promote tRNA recognition. Indeed, one of these mutants (K139A) showed an increase in Km (Table 1), and all four exhibited decreases in kcat, indicating that the transition state of the tRNA-charging reaction was modified.

Archaeal and eukaryal LSDs are not involved in editing

LeuRSs mis-activate a certain number of amino acids that are subsequently charged on tRNA [30,31]. Mis-charged products are edited by a post-transfer editing reaction occurring in a discrete editing domain called CP1. Editing of the wrong amino acid on the terminal adenosine occurs after switching of the tRNA CCA end from the synthetic to the editing site [32]. Previous studies have shown that specific contacts occur during the post-transfer editing reaction in order to stabilize the tRNA in the editing conformation. For instance, the stabilization can involve the anticodon loop of the tRNA [33] or corner of the L-shaped tRNA [34]. The experiments performed in the present study showed that Ph-LSD1 and Gl-LSD1 do not appear to be important to the editing complex formation. The LSD1s were interchangeable without any deleterious effect on the post-transfer editing activities, although the chimaeric proteins lost their synthesis activities. In addition, the mutations of the lysine residues from the Gl-LSD1 that decreased the tRNA-charging activity did not affect the editing activities. Together, these data showed uncoupling of the editing and aminoacylation effects, suggesting that LSD1 is not involved in maintaining the translation fidelity through the editing process.

In conclusion, our data suggested that the LSD1 evolved in order to improve the catalytic properties of LeuRSs. The minimal LSD1, corresponding to the archaeal domain, seems to play a role in the first step of amino acid activation. The eukaryal extension plays a significant role in tRNA charging. Although these results were obtained with PhLeuRS and GlLeuRS as model organisms, the next step towards understanding further the function of LSD1 will require the study of additional mutants of LSD1 from other organisms.

Abbreviations

     
  • aaRS

    aminoacyl-tRNA synthetase

  •  
  • CP

    connective peptide

  •  
  • DTT

    dithiothreitol

  •  
  • Gl

    Giardia lamblia

  •  
  • IleRS

    isoleucyl-tRNA synthetase

  •  
  • LB

    Luria–Bertani

  •  
  • LeuRS

    leucyl-tRNA synthetase

  •  
  • LSD

    leucine-specific domain

  •  
  • Ni-NTA

    Ni2+-nitrilotriacetate

  •  
  • Ph

    Pyrococcus horikoshii

  •  
  • ValRS

    valyl-tRNA synthetase

AUTHOR CONTRIBUTION

En-Duo Wang designed the research. Xiao-Long Zhou, Meng Wang, Min Tan and Qian Huang performed the research. Xiao-Long Zhou, Gilbert Eriani and En-Duo Wang analysed the data. Xiao-Long Zhou, Gilbert Eriani and En-Duo Wang wrote the paper.

We thank Mr Yi-Zu Jiao and Dr Fang Yang for their help during the experiments.

FUNDING

This work was funded by the Natural Science Foundation of China [grant numbers 30670463 and 30930022]; National Key Basic Research Foundation of China [grant number 2006CB910301]; 973 project [grant number 2005CB724600] of China; Committee of Science and Technology in Shanghai [grant number 09JC1415900]; and the Programme International de Coopération Scientifique from Centre National de la Recherche Scientifique [grant number 3606].

References

References
1
Ibba
M.
Söll
D.
Aminoacyl-tRNA synthesis
Annu. Rev. Biochem.
2000
, vol. 
69
 (pg. 
617
-
650
)
2
Ribas de Pouplana
L.
Schimmel
P.
Aminoacyl-tRNA synthetases: potential markers of genetic code development
Trends Biochem. Sci.
2001
, vol. 
26
 (pg. 
591
-
596
)
3
Schimmel
P.
Aminoacyl tRNA synthetases: general scheme of structure–function relationships in the polypeptides and recognition of transfer RNAs
Annu. Rev. Biochem.
1987
, vol. 
56
 (pg. 
125
-
158
)
4
Fersht
A. R.
Editing mechanisms in protein synthesis: rejection of valine by the isoleucyl-tRNA synthetase
Biochemistry
1977
, vol. 
16
 (pg. 
1025
-
1030
)
5
Nureki
O.
Vassylyev
D. G.
Tateno
M.
Shimada
A.
Nakama
T.
Fukai
S.
Konno
M.
Hendrickson
T. L.
Schimmel
P.
Yokoyama
S.
Enzyme structure with two catalytic sites for double-sieve selection of substrate
Science
1998
, vol. 
280
 (pg. 
578
-
582
)
6
Fukai
S.
Nureki
O.
Sekine
S.
Shimada
A.
Tao
J.
Vassylyev
D. G.
Yokoyama
S.
Structural basis for double-sieve discrimination of L-valine from L-isoleucine and L-threonine by the complex of tRNAVal and valyl-tRNA synthetase
Cell
2000
, vol. 
103
 (pg. 
793
-
803
)
7
Chen
J. F.
Guo
N. N.
Li
T.
Wang
E. D.
Wang
Y. L.
CP1 domain in Escherichia coli leucyl-tRNA synthetase is crucial for its editing function
Biochemistry
2000
, vol. 
39
 (pg. 
6726
-
6731
)
8
Eriani
G.
Delarue
M.
Poch
O.
Gangloff
J.
Moras
D.
Partition of tRNA synthetases into two classes based on mutually exclusive sets of sequence motifs
Nature
1990
, vol. 
347
 (pg. 
203
-
206
)
9
Fukunaga
R.
Yokoyama
S.
Crystal structure of leucyl-tRNA synthetase from the archaeon Pyrococcus horikoshii reveals a novel editing domain orientation
J. Mol. Biol.
2005
, vol. 
346
 (pg. 
57
-
71
)
10
Beebe
K.
Merriman
E.
Ribas De Pouplana
L.
Schimmel
P.
A domain for editing by an archaebacterial tRNA synthetase
Proc. Natl. Acad. Sci. U.S.A.
2004
, vol. 
101
 (pg. 
5958
-
5963
)
11
Ling
C.
Yao
Y. N.
Zheng
Y. G.
Wei
H.
Wang
L.
Wu
X. F.
Wang
E. D.
The C-terminal appended domain of human cytosolic leucyl-tRNA synthetase is indispensable in its interaction with arginyl-tRNA synthetase in the multi-tRNA synthetase complex
J. Biol. Chem.
2005
, vol. 
280
 (pg. 
34755
-
34763
)
12
Fukunaga
R.
Yokoyama
S.
Aminoacylation complex structures of leucyl-tRNA synthetase and tRNALeu reveal two modes of discriminator-base recognition
Nat. Struct. Mol. Biol.
2005
, vol. 
12
 (pg. 
915
-
922
)
13
Tukalo
M.
Yaremchuk
A.
Fukunaga
R.
Yokoyama
S.
Cusack
S.
The crystal structure of leucyl-tRNA synthetase complexed with tRNALeu in the post-transfer-editing conformation
Nat. Struct. Mol. Biol.
2005
, vol. 
12
 (pg. 
923
-
930
)
14
Zhou
X. L.
Zhu
B.
Wang
E. D.
The CP2 domain of leucyl-tRNA synthetase is crucial for amino acid activation and post-transfer editing
J. Biol. Chem.
2008
, vol. 
283
 (pg. 
36608
-
36616
)
15
Zheng
Y. G.
Wei
H.
Ling
C.
Martin
F.
Eriani
G.
Wang
E. D.
Two distinct domains of the β subunit of Aquifex aeolicus leucyl-tRNA synthetase are involved in tRNA binding as revealed by a three-hybrid selection
Nucleic Acids Res.
2004
, vol. 
32
 (pg. 
3294
-
3303
)
16
Vu
M. T.
Martinis
S. A.
A unique insert of leucyl-tRNA synthetase is required for aminoacylation and not amino acid editing
Biochemistry
2007
, vol. 
46
 (pg. 
5170
-
5176
)
17
Xu
M. G.
Chen
J. F.
Martin
F.
Zhao
M. W.
Eriani
G.
Wang
E. D.
Leucyl-tRNA synthetase consisting of two subunits from hyperthermophilic bacteria Aquifex aeolicus
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
41590
-
41596
)
18
Ma
J. J.
Zhao
M. W.
Wang
E. D.
Split leucine-specific domain of leucyl-tRNA synthetase from the hyperthermophilic bacterium Aquifex aeolicus
Biochemistry
2006
, vol. 
45
 (pg. 
14809
-
14816
)
19
Schmitt
E.
Meinnel
T.
Blanquet
S.
Mechulam
Y.
Methionyl-tRNA synthetase needs an intact and mobile 332KMSKS336 motif in catalysis of methionyl adenylate formation
J. Mol. Biol.
1994
, vol. 
242
 (pg. 
566
-
576
)
20
Low
B.
Gates
F.
Goldstein
T.
Söll
D.
Isolation and partial characterization of temperature-sensitive Escherichia coli mutants with altered leucyl- and seryl-transfer ribonucleic acid synthetases
J. Bacteriol.
1971
, vol. 
108
 (pg. 
742
-
750
)
21
Zhou
X. L.
Yao
P.
Ruan
L. L.
Zhu
B.
Luo
J.
Qu
L. H.
Wang
E. D.
A unique insertion in the CP1 domain of Giardia lamblia leucyl-tRNA synthetase
Biochemistry
2009
, vol. 
48
 (pg. 
1340
-
1347
)
22
Xu
M. G.
Li
J.
Du
X.
Wang
E. D.
Groups on the side chain of T252 in Escherichia coli leucyl-tRNA synthetase are important for discrimination of amino acids and cell viability
Biochem. Biophys. Res. Commun.
2004
, vol. 
318
 (pg. 
11
-
16
)
23
Fersht
A. R.
Leatherbarrow
R. J.
Wells
T. N.
Structure–activity relationships in engineered proteins: analysis of use of binding energy by linear free energy relationships
Biochemistry
1987
, vol. 
26
 (pg. 
6030
-
6038
)
24
Kobayashi
T.
Takimura
T.
Sekine
R.
Kelly
V. P.
Kamata
K.
Sakamoto
K.
Nishimura
S.
Yokoyama
S.
Structural snapshots of the KMSKS loop rearrangement for amino acid activation by bacterial tyrosyl-tRNA synthetase
J. Mol. Biol.
2005
, vol. 
346
 (pg. 
105
-
117
)
25
Francklyn
C.
Perona
J. J.
Puetz
J.
Hou
Y. M.
Aminoacyl-tRNA synthetases: versatile players in the changing theater of translation
RNA
2002
, vol. 
8
 (pg. 
1363
-
1372
)
26
Zhou
Q.
Kapoor
M.
Guo
M.
Belani
R.
Xu
X.
Kiosses
W. B.
Hanan
M.
Park
C.
Armour
E.
Do
M. H.
, et al. 
Orthogonal use of a human tRNA synthetase active site to achieve multifunctionality
Nat. Struct. Mol. Biol.
2010
, vol. 
17
 (pg. 
57
-
61
)
27
Lee
J. W.
Beebe
K.
Nangle
L. A.
Jang
J.
Longo-Guess
C. M.
Cook
S. A.
Davisson
M. T.
Sundberg
J. P.
Schimmel
P.
Ackerman
S. L.
Editing-defective tRNA synthetase causes protein misfolding and neurodegeneration
Nature
2006
, vol. 
443
 (pg. 
50
-
55
)
28
Nangle
L. A.
Zhang
W.
Xie
W.
Yang
X. L.
Schimmel
P.
Charcot–Marie–Tooth disease-associated mutant tRNA synthetases linked to altered dimer interface and neurite distribution defect
Proc. Natl. Acad. Sci. U.S.A.
2007
, vol. 
104
 (pg. 
11239
-
11244
)
29
Yang
X. L.
Kapoor
M.
Otero
F. J.
Slike
B. M.
Tsuruta
H.
Frausto
R.
Bates
A.
Ewalt
K. L.
Cheresh
D. A.
Schimmel
P.
Gain-of-function mutational activation of human tRNA synthetase procytokine
Chem. Biol.
2007
, vol. 
14
 (pg. 
1323
-
1333
)
30
Martinis
S. A.
Fox
G. E.
Non-standard amino acid recognition by Escherichia coli leucyl-tRNA synthetase
Nucleic Acids Symp. Ser.
1997
, vol. 
36
 (pg. 
125
-
128
)
31
Tan
M.
Zhu
B.
Zhou
X. L.
He
R.
Chen
X.
Eriani
G.
Wang
E. D.
tRNA-dependent pre-transfer editing by prokaryotic leucyl-tRNA synthetase
J. Biol. Chem.
2010
, vol. 
285
 (pg. 
3235
-
3244
)
32
Silvian
L. F.
Wang
J.
Steitz
T. A.
Insights into editing from an Ile-tRNA synthetase structure with tRNAIle and mupirocin
Science
1999
, vol. 
285
 (pg. 
1074
-
1077
)
33
Yao
P.
Zhu
B.
Jaeger
S.
Eriani
G.
Wang
E. D.
Recognition of tRNALeu by Aquifex aeolicus leucyl-tRNA synthetase during the aminoacylation and editing steps
Nucleic Acids Res.
2008
, vol. 
36
 (pg. 
2728
-
2738
)
34
Nordin
B. E.
Schimmel
P.
RNA determinants for translational editing: mischarging a minihelix substrate by a tRNA synthetase
J. Biol. Chem.
1999
, vol. 
274
 (pg. 
6835
-
6838
)

Supplementary data