Nucleotide 34 in tRNA is extensively modified to ensure translational fidelity and efficacy in cells. The deamination of adenosine at this site catalyzed by the enzyme TadA gives rise to inosine (I), which serves as a typical example of the wobble hypothesis due to its diverse basepairing capability. However, recent studies have shown that tRNAArgACG in Mycoplasma capricolum contains unmodified adenosine, in order to decode the CGG codon. The structural basis behind the poorly performing enzyme M. capricolum TadA (McTadA) is largely unclear. Here we present the structures of the WT and a mutant form of McTadA determined at high resolutions. Through structural comparison between McTadA and other active TadA enzymes as well as modeling efforts, we found that McTadA presents multiple structural conflicts with RNA substrates and thus offered support to previous studies from a structural perspective. These clashes would potentially lead to reduced substrate binding affinity of McTadA, consistent with our in vitro deamination activity and binding assays. To rescue the deamination activity of McTadA, we carried out two rounds of protein engineering through structure-guided design. The unsuccessful attempts of the activity restoration could be attributed to the altered dimer interface and stereo hindrance from the non-catalytic subunit of McTadA, which could be the inevitable outcome of the natural evolution. Our study provides structural insight into an alternative decoding and evolutionary strategy by a compromised TadA enzyme at a molecular level.

Introduction

Protein synthesis relies primarily on the continuous interactions between the three major RNAs, i.e. rRNA, tRNA and mRNA, respectively in ribosomes. The production of the polypeptide chain requires the recognition and binding of the codon in mRNA by the anticodon of tRNA [1]. tRNA acts as an adaptor molecule in translation and links the genetic information and the corresponding amino acid. Its maturation process generates a great variety of post-transcriptionally modified nucleotides, and more than a hundred modifications have been found in tRNA to date (reviewed in [2,3]). These modifications are extremely important for the structural stability and the decoding function of tRNA [2,4], as defects in corresponding modification enzymes have often been linked to various human diseases. For example, the knockout of the zebrafish mtu1 gene results in the abolition of the 2-thiouridine modification of U34 in mitochondrial tRNALys, tRNAGlu and tRNAGln. This modification defect inhibits the synthesis of respiratory chain-related proteins, and eventually leads to diseases associated with mitochondrial dysfunction and pathogenesis in the auditory organs [5,6].

The modifications of the nucleotide at position 34 in tRNA are particularly important to all species. Correctly modified tRNA at this site efficiently recognizes and binds the codon triplet on mRNA with enhanced efficiencies [7,8]. tRNA-adenosine deaminase (Tad) or adenosine deaminase acting on tRNA (ADAT) from bacteria and eukaryotes specifically catalyzes the deamination of A34, and this essential modification is conserved in both prokaryotes and eukaryotes [9]. An inosine (I) is subsequently produced and reads the last base of the codon, which could be A, C, U. Therefore, Tad plays an irreplaceable role in the decoding process.

Tads possess the characteristic H(C)XE and PCXXC zinc-binding hallmark sequences, and the zinc ion is critical for the deamination activity [10–13]. Three protein residues (two cysteines and a histidine) and a water molecule are involved in the coordination of Zn2+ [10,11]. The zinc-co-ordinated water molecule is stabilized by the metal and acts as a nucleophile to attack C6 of the target adenosine in a later step [10]. Although employing identical chemistry for catalysis, there are clear differences between the Tad enzymes from eukarya and bacteria, in terms of the deamination process. The bacterial enzyme named TadA usually functions as a homodimer. In contrast, the eukaryotic Tad consists of the heterodimeric Tad2/Tad3 complex [12,13]. Tad2 is the catalytic subunit and Tad3 serves as the regulatory subunit because it lacks a glutamate in the conserved H(C)XE motif, which is replaced by a valine residue. Accordingly, the modification subjects (tRNA species) in the two kingdoms are quite different as well. Sequence alignment indicates that eukaryotic Tads share only limited homology with that of bacteria.

Following the structure determination of numerous bacterial TadA enzymes such as Escherichia coli TadA (EcTadA, PDB 1Z3A), Aquifex aeolicus TadA (AaTadA, PDB 1WWR) [11,14] etc., Losey et al. [10] successfully obtained the structure of Staphylococcus aureus TadA (SaTadA, PDB 2B3J) complexed with the anticodon stem–loop of tRNAArg (Samini-RNA). All these TadAs invariably form homodimers in solution. The determination of the SaTadA-Samini-RNA complex structure provides insight into the molecular mechanism by which TadA specifically recognizes and catalyzes RNA deamination. The asymmetric unit contains two dimers of SaTadA [10]. Both dimers make extensive contacts with the Samini-RNA substrate (especially the five internal bases), with the C32–A38 forming an unusual base pair to cap the end of the loop. In addition, in the active site, the RNA substrate undergoes remarkable structural rearrangements within the loop region (nucleotides 33–37) [10].

Using the bioinformatics approach, Yokobori et al. [15] found that more than 30 species from Mycoplasmas and other Mollicutes were missing the tRNAArg-encoding gene, whose product specifically pairs with the arginine codon CGG. Additionally, 25 such species completely lack the tadA gene, suggesting non-classical decoding strategies in these species. In several species including Mycoplasma capricolum, only tRNAArgACG, tRNAArgTCT and a copy of tadA gene were found to be present in the genome, and the former was presumably to decode the CGN quartet codons. Because inosine (resulted from A34 deamination) is considered as a poor basepairing partner of guanosine due to the unfavorable purine–purine basepairing pattern (except for the A-I pairing) in Crick's theory, it raises the question as to how the CGG codon in these species was decoded. Further investigation of the M. capricolum tRNAArgACG sequence by reverse transcriptase–PCR on extracted tRNAArgACG indicated that some of these tRNA molecules contain un-deaminated A34, as adenosine at the wobble position was still found in 5 out of the 81 clones, while it was zero for B. subtilis [15], implying reduced deamination efficiency of the TadA enzyme in the former species. Moreover, sequence alignment and structural modeling suggested that this observation might be the result of the reduction in deamination efficiency due to selective mutations in TadA during evolution, which interferes with the binding and modification of the tRNA substrates. Consequently, the unmodified tRNAArgACG is responsible for reading the CGG arginine codon [15]. However, their hypothesis was not confirmed due to a lack of structural information on M. capricolum TadA (McTadA).

In the current study, we report the crystal structures of McTadA and its mutants. Based on the structural alignment of McTadA and SaTadA, we designed a series of point mutations to help McTadA to regain the deaminase activity. In addition to multiple possible structural clashes between the enzyme and the RNA substrate, the tRNA-binding cleft of McTadA is found to be wider than that of other bacterial TadA enzymes, which may also contribute to the inability of the enzyme to efficiently interact with and deaminate tRNAs. In summary, our study on McTadA helps to explore the evolutionary scenario of the decoding process in Mycoplasmas and other Mollicutes, and aids in the understanding of the adenosine deaminase and the CDA superfamily from the structural and evolutionary perspectives.

Materials and methods

Cloning, overexpression, protein purification

The gene encoding M. capricolum tadA (accession ABC01127) was chemically synthesized, which was digested by NdeI and XhoI and ligated into the vector pET-21b (+) (designated as TadA-pET21b). The construct contains the full-length gene fused with a C-terminal 6 × His tag. The vector was then transformed into the Transetta (DE3) strain (Transgen Biotech). A 20-ml overnight culture was used to inoculate 2 L of LB supplemented with 30 µg/ml kanamycin, 64 µg/ml chloramphenicol and 100 mM ZnCl2. Cells were grown to an OD600 of 0.6–0.8 at 37°C, at which point the temperature was lowered to 16°C, and induced with 0.5 mM IPTG. Overnight cells were harvested and sonicated in the lysis buffer (20 mM Tris–HCl pH 8.0, 250 mM NaCl, 20 mM imidazole, 10 mM β-mercaptoethanol (β-ME), 15 µM ZnCl2 and 15% glycerol). The lysate was clarified by centrifugation at 23500g for 60 min at 4°C. The resulting supernatant was incubated with 1–2 ml of Ni-NTA agarose (Qiagen) equilibrated with buffer A (20 mM Tris–HCl pH 8.0, 250 mM NaCl, 20 mM imidazole, 10 mM β-ME, 15 µM ZnCl2 and 15% glycerol) for 60 min. Gradual removal of indigenous proteins was carried out using an imidazole gradient. Buffer A supplemented with 250 mM imidazole was used to elute the TadA protein. The purity of the target protein was more than 90% as judged by SDS–PAGE, but degradation was observed at room temperature. TadA-containing fractions were pooled and diluted two-fold with buffer A. Next the protein was loaded onto a 5-ml HiTrap Heparin column (GE Healthcare) and eluted by a NaCl gradient supplemented with 10% glycerol. The eluate was dialyzed to a buffer containing 20 mM Tris–HCl pH 8.0, 250 mM NaCl, 2 mM PMSF, 10 mM β-ME and 15 µM ZnCl2. Fractions were pooled and eventually concentrated to 2–3 mg/ml, with a yield of ∼0.5 mg/L medium.

The genes for TadA from Bacillus subtilis (BsTadA), Staphylococcus aureus (SaTadA) and Escherichia coli (EcTadA) were chemically synthesized, which were digested by NdeI and XhoI and ligated into the pET-21b (+) or pET-28a (+) vector (Merck). The vectors were then transformed into the Transetta (DE3) strain (Transgen Biotech). The yeast Tad2 and Tad3 genes were amplified by PCR from S. cerevisiae cDNA using the primers Tad2-F1/R1 and Tad3-F1/R1, respectively. After double digestion by corresponding restriction enzymes, the two genes were sequentially ligated into the multiple cloning site (MCS) of the pETDuet-1 vector (Merck). After Tad3 was inserted into MCS1 using the EcoRI and HindIII restriction sites, Tad2 was inserted into MCS2 of the same vector using the BglII and XhoI sites. Then the entire DNA fragment carrying both genes was further amplified from the vector using the primers Tad3-F2 and Tad2-R1, and subcloned into pET-28a (+) vector (Merck). All the primers used in this study were listed in Table 1. The overexpression, purification and follow-up treatment of the target proteins followed the same protocol as that of the WT McTadA protein.

Table 1
Sequences of the primers and tRNA used in this study
Primers nameSequence (5′–3′)
TadA-pET28a-F CGGCAGCCATATGGACGACTTTAACAATATC 
TadA-pET28a-R GGCCGCACTCGAGTTAACGTTTGTTGATGAAGAA 
TadA-pCold-MBP-F GGCCGCACTCGAGATGGACGACTTTAACAATATC 
TadA-pCold-MBP-R CCGCGCAAAGCTTTTAACGTTTGTTGATGAAGAA 
TadA-pET21b-F CGGCAGCCATATGGACGACTTTAACAATATC 
TadA-pET21b-R GGCCGCACTCGAGACGTTTGTTGATGAAGAA 
S105-I110(DPK)-pET21b-F CCGAAGAAAAACAACTATAGCATC 
S105-I110(DPK)-pET21b-R GTCATCAACCAGGTAGTAGAT 
I19-N22 (AQLG)-pET21b-F CTTGGAGACATTCCGGTGAGCTGC 
I19-N22 (AQLG)-pET21b-R CTGGGCTGCCTTTTTACTTTCGTTAATC 
K146A-pET21b-F ATCTTCTTCATCAACGCACGTCTCGAGCACCAC 
K146A-pET21b-R GTGGTGCTCGAGACGTGCGTTGATGAAGAAGAT 
R45P-Y46I-pET21b-F ATTAAAAACAAAGATATCAGCCAG 
R45P-Y46I-pET21b-R GGGGCTATTAATTGCCAGGCTCAG 
EcTadA-E59Q-pET28a-F CCCACCGCACATGCACAAATCATGGCCCTGCGGC 
EcTadA-E59Q-pET28a-R GCCGCAGGGCCATGATTTGTGCATGTGCGGTGGG 
McTadA-E56Q-pET21b-F ATCAGCCAGCATGCACAAATTAACGTCATCAATG 
McTadA-E56Q-pET21b-R CATTGATGACGTTAATTTGTGCATGCTGGCTGAT 
SaTadA-E55Q-pET28a-F CCGACCGCACATGCACAACATATTGCAATCGAA 
SaTadA-E55Q-pET28a-R TTCGATTGCAATATGTTGTGCATGTGCGGTCGG 
BsTadA-E55Q-pET21b-F TCTATTGCCCATGCACAAATGCTTGTAATTG 
BsTadA-E55Q-pET21b-R CAATTACAAGCATTTGTGCATGGGCAATAGA 
Tad3-F1 GCTTAGGAATTCGATGGTTAAGAAAGTT 
Tad3-R1 ATAGGAAAGCTTTCAGCAGCAGACATCCCG 
Tad2-F1 CATATGGCAGATCTCCAGCATATTAAACAT 
Tad2-R1 TTGCACTTCTCGAGTCAAACCTTAGATTTCCTTATG 
Tad3-F2 ATAGGAGCTAGCATGGTTAAGAAAGTTAATAAT 
MctRNAArg GCGCCCGTAGATCAATTGGATAGATCGCTTGACTAC
GGATCAAAAGGTTGGGGGTTCGAGTCCCTCCGGGC
GCACCA 
Samini-RNA GGUGACUACGGAUCACC 
Primers nameSequence (5′–3′)
TadA-pET28a-F CGGCAGCCATATGGACGACTTTAACAATATC 
TadA-pET28a-R GGCCGCACTCGAGTTAACGTTTGTTGATGAAGAA 
TadA-pCold-MBP-F GGCCGCACTCGAGATGGACGACTTTAACAATATC 
TadA-pCold-MBP-R CCGCGCAAAGCTTTTAACGTTTGTTGATGAAGAA 
TadA-pET21b-F CGGCAGCCATATGGACGACTTTAACAATATC 
TadA-pET21b-R GGCCGCACTCGAGACGTTTGTTGATGAAGAA 
S105-I110(DPK)-pET21b-F CCGAAGAAAAACAACTATAGCATC 
S105-I110(DPK)-pET21b-R GTCATCAACCAGGTAGTAGAT 
I19-N22 (AQLG)-pET21b-F CTTGGAGACATTCCGGTGAGCTGC 
I19-N22 (AQLG)-pET21b-R CTGGGCTGCCTTTTTACTTTCGTTAATC 
K146A-pET21b-F ATCTTCTTCATCAACGCACGTCTCGAGCACCAC 
K146A-pET21b-R GTGGTGCTCGAGACGTGCGTTGATGAAGAAGAT 
R45P-Y46I-pET21b-F ATTAAAAACAAAGATATCAGCCAG 
R45P-Y46I-pET21b-R GGGGCTATTAATTGCCAGGCTCAG 
EcTadA-E59Q-pET28a-F CCCACCGCACATGCACAAATCATGGCCCTGCGGC 
EcTadA-E59Q-pET28a-R GCCGCAGGGCCATGATTTGTGCATGTGCGGTGGG 
McTadA-E56Q-pET21b-F ATCAGCCAGCATGCACAAATTAACGTCATCAATG 
McTadA-E56Q-pET21b-R CATTGATGACGTTAATTTGTGCATGCTGGCTGAT 
SaTadA-E55Q-pET28a-F CCGACCGCACATGCACAACATATTGCAATCGAA 
SaTadA-E55Q-pET28a-R TTCGATTGCAATATGTTGTGCATGTGCGGTCGG 
BsTadA-E55Q-pET21b-F TCTATTGCCCATGCACAAATGCTTGTAATTG 
BsTadA-E55Q-pET21b-R CAATTACAAGCATTTGTGCATGGGCAATAGA 
Tad3-F1 GCTTAGGAATTCGATGGTTAAGAAAGTT 
Tad3-R1 ATAGGAAAGCTTTCAGCAGCAGACATCCCG 
Tad2-F1 CATATGGCAGATCTCCAGCATATTAAACAT 
Tad2-R1 TTGCACTTCTCGAGTCAAACCTTAGATTTCCTTATG 
Tad3-F2 ATAGGAGCTAGCATGGTTAAGAAAGTTAATAAT 
MctRNAArg GCGCCCGTAGATCAATTGGATAGATCGCTTGACTAC
GGATCAAAAGGTTGGGGGTTCGAGTCCCTCCGGGC
GCACCA 
Samini-RNA GGUGACUACGGAUCACC 

The restriction sites of the cloning primers were underlined in the amplifying primers while the mutated bases were in italic in the Quikchange primers.

Point mutation and protein engineering

Site-directed mutagenesis was performed using the Fast Mutagenesis System (Transgen Biotech) according to the manufacturer's instructions. Primers for mutation of specific amino acids were shown in Table 1. The WT plasmids were used as the templates to obtain corresponding mutants for overexpression in Transetta (DE3). The purification method of the mutant protein follows the protocol of the WT proteins.

Crystallization, data collection and structure determination

An amount of 2.5 mg/ml TadA and 10 mM β-ME, 15 µM ZnCl2 were mixed and incubated on ice for 40 min, and the sample was then mixed with crystallization well solutions at a ratio of 1 : 1 (v/v). The crystals of the WT McTadA were grown at 4°C using the hanging drop vapor diffusion method and the best crystallization condition was 0.49 M sodium phosphate monobasic monohydrate, 0.91 M potassium phosphate dibasic, pH 6.9. The crystals of McTadA-TM were obtained under the condition of 1.4 M ammonium sulfate, 0.1 M sodium citrate pH 6.0. Crystals grew to full size over 3 days at 4°C. Before being flash-cooled, crystals were transferred into the cryo-protective solution composed of the components from the well solution and glycerol of increasing concentrations, to a final concentration of 20% (v/v). Crystals were then mounted on a nylon loop and flash-frozen in liquid nitrogen. Data collection of the WT crystals was carried out on an in-house XtaLab Synergy single crystal diffractometer (Oxford Diffraction), operating at 40 kV and 30 mA with a rotating copper-anode generator of a wavelength of 1.54 Å. The diffraction dataset was recorded with a 51.5-mm Onyx CCD detector. Data reduction was performed using the CrysAlisPro software (Oxford Diffraction). The McTadA-TM data were collected at Beamline 19U (BL19U1) at the Shanghai Synchrotron Radiation Facility (SSRF) with a wavelength of 0.979 Å, which was subsequently processed with the program HKL3000 [16]. All data were collected at 100 K with a nitrogen cryo stream.

The WT crystals belonged to the P43212 space group with 2.4 Å resolution and the asymmetric unit was predicted to contain one monomer. Molecular replacement (MR) was first performed with Phenix using the AfTadA (PDB 2A8N) as the search model [17]. After a plausible solution was obtained, the model was manually rebuilt by COOT according to the electron density map [18,19]. The rebuilt model was fed to the phenix.refine and multiple cycles of refinement were conducted, followed by model rebuilding [20,21]. The Rfree/Rwork factors were 21.9- and 19.8%, respectively and the final model was validated by Molprobity [22]. The structure of the TM mutant was solved by MR using the WT structure as the model. The Rfree/Rwork factors for the mutant were 25.1- and 19.8%, respectively. The structural presentation of McTadA was prepared by PyMOL (http://www.pymol.org/). All data collection and refinement statistics are presented in Table 2.

Table 2
Data collection and refinement statistics
Apo-WT (6L2L)TM (6L2M)
Data collection Home X-ray SSRF-BL19U 
 Resolution (Å) 21.10–2.40 (2.53–2.40)a 50.00–2.30 (2.38–2.30) 
 Space group P4321P3221 
 Cell dimensions(Å) 
  a, b, c (Å) 49.1, 49.1, 147.9 55.6, 55.6, 109.2 
  α, β, γ (°) 90, 90, 90 90, 90, 90 
Rmergeb 0.10 (0.46) 0.105 (0.51) 
  CC1/2 0.998 (0.884) 0.990 (0.969) 
  Redundancy 6.6 (7.1) 18.7 (18.1) 
 Completeness (%) 99.8 (99.9) 100 (100) 
 I/σ(I) 15.6 (4.5) 38.5 (4.8) 
Refinement 
 Resolution (Å) 21.10–2.40 (2.75–2.40) 48.15–2.30 (2.64–2.30) 
 No. reflections 7649 9119 
Rworkc/Rfreed(%) 19.8/21.9 19.8/25.1 
 No. atoms 
  Protein 1160 1147 
  Ligand 1 (Zn2+1 (Zn2+), 2(Cl
  Water 59 38 
 B-factor (Å2
  Protein 36.7 48.1 
  Ligand 27.9 31.3 (Zn2+), 54.8 (Cl
  Water 32.5 46.0 
  RMS (bonds) (Å) 0.005 0.007 
  RMS (angles) (°) 0.81 0.81 
 Ramachandran favored (%) 96.40 95.71 
 Outliers (%) 
Apo-WT (6L2L)TM (6L2M)
Data collection Home X-ray SSRF-BL19U 
 Resolution (Å) 21.10–2.40 (2.53–2.40)a 50.00–2.30 (2.38–2.30) 
 Space group P4321P3221 
 Cell dimensions(Å) 
  a, b, c (Å) 49.1, 49.1, 147.9 55.6, 55.6, 109.2 
  α, β, γ (°) 90, 90, 90 90, 90, 90 
Rmergeb 0.10 (0.46) 0.105 (0.51) 
  CC1/2 0.998 (0.884) 0.990 (0.969) 
  Redundancy 6.6 (7.1) 18.7 (18.1) 
 Completeness (%) 99.8 (99.9) 100 (100) 
 I/σ(I) 15.6 (4.5) 38.5 (4.8) 
Refinement 
 Resolution (Å) 21.10–2.40 (2.75–2.40) 48.15–2.30 (2.64–2.30) 
 No. reflections 7649 9119 
Rworkc/Rfreed(%) 19.8/21.9 19.8/25.1 
 No. atoms 
  Protein 1160 1147 
  Ligand 1 (Zn2+1 (Zn2+), 2(Cl
  Water 59 38 
 B-factor (Å2
  Protein 36.7 48.1 
  Ligand 27.9 31.3 (Zn2+), 54.8 (Cl
  Water 32.5 46.0 
  RMS (bonds) (Å) 0.005 0.007 
  RMS (angles) (°) 0.81 0.81 
 Ramachandran favored (%) 96.40 95.71 
 Outliers (%) 
a

Values in parentheses are for the highest-resolution shell.

b

Rmerge = Σ|(I − )|/σ(I), where I is the observed intensity.

c

Rwork = Σhkl||Fo| − |Fc||/Σhkl|Fo|, calculated from working data set.

d

Rfree is calculated from 5.0% of data randomly chosen and not included in refinement.

In vitro transcription of RNA

The MctRNAArg was prepared by in vitro transcription [23]. The transcribed RNA was purified by 12% denaturing Urea-PAGE gel, extracted, and was then ethanol-precipitated. The pellet was washed and redissolved in the TE buffer (10 mM Tris–HCl pH 8.0 and 1 mM EDTA) to a concentration of 5 mg/ml and annealed by heating to 65°C and allowed to cool to room temperature. The annealed RNA was stored at −80°C for further use. The Samini-RNA based on the anticodon stem–loop sequence of SatRNAArg for the ITC assay was chemically synthesized.

Coupled deaminase activity assays

An amount of 2.5 µM MctRNAArg or Samini-RNA was incubated with 2.5 µM McTadA (WT or McTadA-TM) in a reaction buffer of 10 mM Tris–HCl pH 8.0, 25 mM KCl, 2.5 mM MgCl2 and 1 mM DTT at 30°C for 60 min. 50 mM CHES pH 9.5, and 2 µM EcEndoV [24–27] were added to this mixture, and the reaction was incubated at 37°C for another 50 min. The reaction was stopped by adding 2× loading buffer. The product was analyzed by 18% urea-PAGE gel electrophoresis followed by ethidium bromide staining and UV-light exposure.

Circular dichroism (CD) analysis

The WT and mutant proteins were first diluted to a concentration of 0.3–0.5 mg/ml in the buffer containing 20 mM Tris-H2SO4, 250 mM Na2SO4, and 1 mM DTT. Far-UV CD measurements were performed on J-810 spectropolarimeter (Jasco, Japan) using a Teflon-sealed quartz glass cuvette (1-mm path length, Starna, U.K.). A wavelength increment of 1 nm, a response time of 4 s, and a scan speed of 20 nm/min were employed during the measurements. All the resulting spectra were buffer-corrected.

Isothermal titration calorimetry (ITC)

The ITC experiments were conducted at 25°C using a PEAQ ITC titration calorimeter (Malvern instruments). To exactly match their buffer compositions, the Samini-RNA and TadA proteins (WT McTadA or BsTadA/E55Q) were dialyzed against the same buffer containing 20 mM Tris–HCl pH 8.0, 150 mM NaCl. The affinities were measured by titrating Samini-RNA (200–300 µM) into WT McTadA or BsTadA/E55Q (30–35 µM). The first injection of 0.4 µl was followed by 25 injections of 1.5 µl drops. The MICROCAL ORIGIN software was used to determine the site-binding models that produced good fits. Individual peaks from the titrations were integrated and displayed on a Wiseman plot. The first reading was removed from the analysis. The dissociative constant (KD) and change in enthalpy (ΔH) associated with the binding events were calculated after fitting the dataset.

Electrophoretic mobility shift assay (EMSA)

An amount of 2 µM tRNA was incubated with McTadA (WT or McTadA-TM) at different molar ratios in a buffer of 20 mM Tris–HCl pH 8.0 and 1 mM DTT on ice for 60 min, and EcTadA, BsTadA, SaTadA and ScTad2/3 were used as controls. The reaction mixture was then mixed with glycerol (10% v/v) before being loaded and analyzed by 6% native gels with a running buffer containing 40 mM Tris–HCl pH 9.2. The gels were electrophoresed at 100 V for 60 min and stained either with ethidium bromide.

Results

The in vitro deamination activity of McTadA

To test the deamination activity toward the ACG codon-containing RNA substrates, we first expressed the full-length recombinant McTadA. The protein was prone to precipitation upon isolation, and had to be stabilized by 10–15% glycerol. The deamination assays were then performed to check McTadA's activity, using the Samini-RNA fragment or the full-length tRNAArg of M. capricolum (MctRNAArg) (Figure 1A). This was a coupled assay utilizing the combined actions of McTadA for deamination and E. coli Endonuclease V (EcEndoV) for RNA cleavage. The latter specifically recognizes inosine (I) and cleaves the second phophodiester bond after I produced by the A34 deamination. The results showed that the enzyme was inactive on both RNA molecules, while the positive control EcTadA in conjunction with EcEndoV gave rise to cleaved bands from both substrates (Figure 1B). Additionally, the incubation of McTadA or EcTadA with the RNA substrates alone did not generate any cleaved products, meaning that there was no unwanted nuclease contamination. We also supplemented Ca2+, Mg2+, Zn2+, or Mn2+ ions at different concentrations separately to the assays, but obtained the same results (Figure 1C). ICP-MS spectrometry demonstrated that the protein contained the Zn2+ ion (data not shown), which ruled out the mis-folding possibility due to its tendency to precipitation. Of note, the addition of the Mn2+ ion resulted in protein precipitation immediately and a smear was seen in the gel. On the other hand, the EcTadA result was interesting, because tRNAArg of E. coli (EctRNAArg) differs in anticodon loop sequence from MctRNAArg or the tRNAArg of Staphylococcus aureus (SatRNAArg), which suggested its broad substrate specificity.

Biochemical characterization of WT McTadA.

Figure 1.
Biochemical characterization of WT McTadA.

(A) The RNA substrates used in the assays. A34 was colored in red. (B) The coupled deaminase activity assay on McTadA. The red arrow indicates the cleaved product at correct size from MctRNAArg. (C) The activity assays with supplemented metals. Two concentrations were employed for each metal.

Figure 1.
Biochemical characterization of WT McTadA.

(A) The RNA substrates used in the assays. A34 was colored in red. (B) The coupled deaminase activity assay on McTadA. The red arrow indicates the cleaved product at correct size from MctRNAArg. (C) The activity assays with supplemented metals. Two concentrations were employed for each metal.

The overall structure of M. capricolum TadA

To explore the underlying structural basis behind McTadA's poor binding affinity, we crystalized the protein and solved the structure at a resolution of 2.4 Å. The crystal was in P43212 crystallographic symmetry and there was only one molecule in the asymmetric unit (Figure 2A). The final structure displayed a good geometry with the Rwork and Rfree values being refined to 19.8% and 21.9%, respectively, and no residues fell in disallowed region in the Ramachrdran plot. The protein displayed a typical tRNA adenosine deaminase fold, with two long helices flanking each side of the central beta-sheet. The sheet is composed of five strands, with β1 antiparallel to all the others. The N- (Asn5-His21) and C-termini (Gln129-Arg147) each forms a long helix, and they run in parallel direction, but do not form any inter-helical contacts. The density from Ser115 to Asp118 was rather poor, and was considered a disorder and not modeled. Otherwise, residues from Asp2 to Leu148 were completely resolved. In accordance to the ICP-MS result, there is a zinc ion bound to the catalytic center. The zinc ion is fixed by four ligands, comprising His54, Cys84, Cys87, and a water molecule. The four ligands form a tetrahedral geometry, with the water molecule held in position by the general base Glu56, whose OE2 atom is only 2.9 Å away (Figure 2B). Additionally, Phe108 from the β4–β5 loop forms possible hydrophobic interactions with Phe143, Leu139, Ile136, all of which are located on the same side of the final helix (Figure 2C). These interactions suggest that there might be cross-talks between these two regions. Crystallographic symmetry operations could generate the enzyme dimer, i.e. the functional catalytic unit, with the two-fold axis coinciding with the dimer interface (Figure 2A).

The overall structure of apo-McTadA.

Figure 2.
The overall structure of apo-McTadA.

(A) The structure was shown in the ribbon rendition in two orthogonal views. The N- and C-termini were indicated and the secondary structure elements of McTadA were labeled. The blue dot indicated the zinc ion. Only one monomer was drawn before the rotation, which was colored according to its secondary structure. The right panel: the protein was drawn in the dimeric form, where the second monomer was shown in gray. (B) The close-up view of the zinc-coordination state. The side chains of the key residues were shown as sticks and labeled. The water molecule for catalysis (Wat) was represented by the pink sphere and its distance from the general base Glu56 was indicated. The distances of the ligands from the metal were shown (units in Å). (C) The hydrophobic interactions of Phe107 with residues Ile136, Leu139 and Phe143 from the terminal helix.

Figure 2.
The overall structure of apo-McTadA.

(A) The structure was shown in the ribbon rendition in two orthogonal views. The N- and C-termini were indicated and the secondary structure elements of McTadA were labeled. The blue dot indicated the zinc ion. Only one monomer was drawn before the rotation, which was colored according to its secondary structure. The right panel: the protein was drawn in the dimeric form, where the second monomer was shown in gray. (B) The close-up view of the zinc-coordination state. The side chains of the key residues were shown as sticks and labeled. The water molecule for catalysis (Wat) was represented by the pink sphere and its distance from the general base Glu56 was indicated. The distances of the ligands from the metal were shown (units in Å). (C) The hydrophobic interactions of Phe107 with residues Ile136, Leu139 and Phe143 from the terminal helix.

Comparison to other TadAs and generation of a McTadA-RNA complex model

A Dali search showed that the closest homologs to McTadA are SaTadA in complex with Samini-RNA (PDB 2B3J) [10], AaTadA (PDB 1WWR) [11], and TadA from Salmonella enterica (SeTadA, PDB 3OCQ), respectively. SaTadA shares moderate sequence identity (26%) and structural similarity with McTadA (an RMSD of 1.6 Å over 135 residues), indicating considerable structural deviation from the latter (Figure 3A,B).

Structure and sequence comparison to homologs.

Figure 3.
Structure and sequence comparison to homologs.

(A) The structural superposition of the apo-McTadA (PDB 6L2L, color scheme as in Figure 1A) with that of tRNA-bound SaTadA (PDB 2B3J, cyan, RNA was omitted for clarity), AaTadA (PDB 1WWR, magenta), and SeTadA (PDB 3OCQ, yellow), respectively. The proteins were represented in backbone traces and in cross-eyed stereoview. (B) Multiple sequence alignments of the structural homologs. The secondary structure elements were labeled above the sequences. The red dots indicated key residues responsible for zinc binding, and the up-arrow indicated the catalytic Glu56 residue. The regions for the first- and second-round engineering were represented by the black and green broken boxes, respectively.

Figure 3.
Structure and sequence comparison to homologs.

(A) The structural superposition of the apo-McTadA (PDB 6L2L, color scheme as in Figure 1A) with that of tRNA-bound SaTadA (PDB 2B3J, cyan, RNA was omitted for clarity), AaTadA (PDB 1WWR, magenta), and SeTadA (PDB 3OCQ, yellow), respectively. The proteins were represented in backbone traces and in cross-eyed stereoview. (B) Multiple sequence alignments of the structural homologs. The secondary structure elements were labeled above the sequences. The red dots indicated key residues responsible for zinc binding, and the up-arrow indicated the catalytic Glu56 residue. The regions for the first- and second-round engineering were represented by the black and green broken boxes, respectively.

By superimposing with SaTadA complexed with Samini-RNA (PDB 2B3J) [10], we could generate the structure of the McTadA-RNA complex (Figure 4A). While the general fold and the zinc ion position are conserved, inspection and analyses of the structures revealed the structural uniqueness for McTadA. Clashes with RNA at two regions are evident, which coincide with the unusual structural features of McTadA: (1) the Ser105-Ile110 fragment would severely clash with U33; (2) the Ile19-Asn22 fragment would bump into C35; and (3) K146 may be in close distance from C35 (3.1 Å, Figure 4B,C). These findings were consistent with the prediction made by Yokobori et al. [15], based on the homology modeling of the SaTadA-Samini-RNA complex (PDB 2B3J) [10]. Therefore, the poor activity of McTadA toward RNA is likely the result of steric hindrance between the enzyme and substrates.

Modeling studies showing potential steric hindrance between McTadA and the Samini-RNA substrate.

Figure 4.
Modeling studies showing potential steric hindrance between McTadA and the Samini-RNA substrate.

(A) Cartoon showing the overall hypothetical McTadA-Samini-RNA complex. The co-ordinates of Samini-RNA (in orange) were from PDB 2B3J and only nucleotides U33–C35 were drawn. The two most severe structural conflicts were indicated by the broken red circles. P5P34: the nebularine base. (B) The close-up of the clashes posed by the Ser105-Ile110 fragment; and (C) by the Ile19-Asn22 fragment as well as Lys146. The key residues were shown as sticks.

Figure 4.
Modeling studies showing potential steric hindrance between McTadA and the Samini-RNA substrate.

(A) Cartoon showing the overall hypothetical McTadA-Samini-RNA complex. The co-ordinates of Samini-RNA (in orange) were from PDB 2B3J and only nucleotides U33–C35 were drawn. The two most severe structural conflicts were indicated by the broken red circles. P5P34: the nebularine base. (B) The close-up of the clashes posed by the Ser105-Ile110 fragment; and (C) by the Ile19-Asn22 fragment as well as Lys146. The key residues were shown as sticks.

Attempts to restore the deamination activity and possible causes of failure

As an attempt to restore the activity of McTadA, we introduced structure-based mutations to the clashed regions: (1) replacement of the Ser105-Ile110 fragment by the Asp-Pro-Lys tri-peptide; (2) substitution of the Ile19-Asn22 fragment by the Ala-Gln-Leu-Gly tetra-peptide; (3) the K146A mutation. These mutations were either based on the SaTadA sequence, or simply reduced the side chain sizes of the clashing residues (in the K146A case). The mutant, which we name McTadA-TM, was even more unstable than WT. Nevertheless, we managed to obtain the pure protein for both biochemical characterization and follow-up crystallization trials. The McTadA-TM mutant was successfully crystallized based on the WT crystallization condition, and the structure was determined at 2.3 Å. The mutant structure resembles that of the WT, with only an RMSD value of 0.56 Å for 133 Cαs. There was still one monomer in the ASU, but the space group changed to P3221. The structural superposition indicated that the key regions of the mutant were rearranged as we have designed, and would remarkably reduce its possible clashes with Samini-RNA, according to our model. Specifically, the previous Ile19-Asn22 fragment was originally located at the top portion of the α1-helix, and after the mutation the replacement fragment rotates to a certain degree to avoid the clashes with C35 (Figure 5A). The K146A mutation considerably reduces the size of the side chain, and now poses little hindrance to C35 as well. Lastly, the substitution at the Ser105-Ile110 region produced local disorder between Lys107-Tyr111 (new numbering according to the TM mutant) and pulled the fragment away from U33. Additionally, the invisible density suggested flexibility of this region, which may help to relieve the rigidity and structural clashes (Figure 5B). From our model, the current structure of the mutant would not interfere with the binding of RNA at these two sites.

Modeling studies showing the relief of the potential steric hindrance between the McTadA-TM mutant and the Samini-RNA substrate (from PDB 2B3J).

Figure 5.
Modeling studies showing the relief of the potential steric hindrance between the McTadA-TM mutant and the Samini-RNA substrate (from PDB 2B3J).

The orientation of each panel was the same as that in Figure 4. (A) Cartoon showing the close-up view of the structural rearrangements by the AQLG fragment (the replacement of the original IKHN fragment at positions of 19–22); and (B) by the 105DPK107 fragment as well as Ala146. The key residues were shown as sticks.

Figure 5.
Modeling studies showing the relief of the potential steric hindrance between the McTadA-TM mutant and the Samini-RNA substrate (from PDB 2B3J).

The orientation of each panel was the same as that in Figure 4. (A) Cartoon showing the close-up view of the structural rearrangements by the AQLG fragment (the replacement of the original IKHN fragment at positions of 19–22); and (B) by the 105DPK107 fragment as well as Ala146. The key residues were shown as sticks.

With these desired engineering purposes presumably accomplished, we next performed activity assays as described above. However, we still could not detect the deamination activity for the McTadA-TM. Circular dichroism studies showed that the TM mutant had quite similar profiles to that of WT, suggesting little structural perturbation by the mutations (Supplementary Figure S1). Thorough structural analyses revealed that the possible reasons might be due to the fact that only the conflicts from one monomer of McTadA were considered when we design the McTadA-TM mutant, but we failed to take account for the possible influence of the other monomer. Moreover, the PDBePISA server [28] indicated that when WT McTadA forms a dimer, it forms quite a wide interfacial cleft (Figure 6A). The buried surface area of McTadA is only 1890 Å2, which is considerably smaller than that of EcTadA (2870 Å2 based on PDB 1Z3A) [14], AaTadA (2800 Å2 based on PDB 1WWR) [11], or Agrobacterium fabrum tRNA adenosine deaminase TadA (AfTadA, 2620 Å2 based on PDB 2A8N) [17]. When RNA substrates bind, they would make contacts with both enzyme subunits. While the anticodon loop bases (especially A34) primarily interact with the catalytic subunit, other bases also require support from the non-catalytic subunit. A larger interfacial area suggests a weaker association between the enzyme and RNA substrates. This is probably caused by the translation of α4 (Ala55-Leu67), which moves away from the dimer interface (Figure 6A). Additionally, close examination of the McTadA dimer further revealed that the R45Y46 fragment from the catalytic subunit and Ser69 from the non-catalytic subunit would block bases C35 and G37, respectively (Figure 6B,C). Of note, the Cα of Ser69 is only 2.1 Å from N1 of G37, a backbone conflict that could not be resolved by simple point mutations. Furthermore, the quintuple mutant (double mutation R45P/Y46I on top of the triple mutant) failed to express in E. coli and we therefore stopped our engineering effort on this aspect. Taken together, the inactivity might still stem from the unresolved steric clashes due to unusual structural features of the McTadA enzyme, where the non-catalytic subunit further reduces the binding possibility of RNA substrates.

Further explanation of the inability of the McTadA-TM mutant to catalyze.

Figure 6.
Further explanation of the inability of the McTadA-TM mutant to catalyze.

(A) The superposition of chain A of the McTadA dimer (with chain A and B colored in cyan and gray, respectively) with that of the EcTadA (PDB 1Z3A, blue), AaTadA (PDB 1WWR, magenta) and AfTadA (PDB 2A8N, orange) dimers in the ribbon representation. Note the translation of α4 as indicated by the arrow. (B) The modeled structure of RNA-bound McTadA dimer, showing more structural clashes between the RNA substrate and the McTadA-TM mutant. Unresolved structural conflicts were shown by the red box. (C) The close-up view showing the proximity of C35 and G37, to Tyr46 of the catalytic subunit and Ser46 from the non-catalytic subunit, respectively. (D) The multiple sequence alignment with a focus on the region of Asn22 and Phe108, indicated by the red arrow head and red arrow, respectively. The bacteria and Mollicute species were divided by the red horizontal line. Nitrosomonas, Nitrosomonas europaea; Aquifex, A. aeolicus; Streptomyces, Streptomyces avermitilis; Synechococcus, Synechococcus elongates; Staphylococcus, S. aureus; Listeria, Listeria monocytogenes; Oenococcus, Oenococcus oeni; Acholeplasma, A. laidlawii; Aster-yellow, Aster yellows witches’-broom phytoplasma; Phyto-aust, Candidatus Phytoplasma australiense; Phyto-mali, Candidatus Phytoplasma mali; Onion-yellow, Onion yellows phytoplasma; Mesoplasma, Mesoplasma florum L1. The secondary structures for EcTadA and McTadA were labeled at the top and bottom of the alignment.

Figure 6.
Further explanation of the inability of the McTadA-TM mutant to catalyze.

(A) The superposition of chain A of the McTadA dimer (with chain A and B colored in cyan and gray, respectively) with that of the EcTadA (PDB 1Z3A, blue), AaTadA (PDB 1WWR, magenta) and AfTadA (PDB 2A8N, orange) dimers in the ribbon representation. Note the translation of α4 as indicated by the arrow. (B) The modeled structure of RNA-bound McTadA dimer, showing more structural clashes between the RNA substrate and the McTadA-TM mutant. Unresolved structural conflicts were shown by the red box. (C) The close-up view showing the proximity of C35 and G37, to Tyr46 of the catalytic subunit and Ser46 from the non-catalytic subunit, respectively. (D) The multiple sequence alignment with a focus on the region of Asn22 and Phe108, indicated by the red arrow head and red arrow, respectively. The bacteria and Mollicute species were divided by the red horizontal line. Nitrosomonas, Nitrosomonas europaea; Aquifex, A. aeolicus; Streptomyces, Streptomyces avermitilis; Synechococcus, Synechococcus elongates; Staphylococcus, S. aureus; Listeria, Listeria monocytogenes; Oenococcus, Oenococcus oeni; Acholeplasma, A. laidlawii; Aster-yellow, Aster yellows witches’-broom phytoplasma; Phyto-aust, Candidatus Phytoplasma australiense; Phyto-mali, Candidatus Phytoplasma mali; Onion-yellow, Onion yellows phytoplasma; Mesoplasma, Mesoplasma florum L1. The secondary structures for EcTadA and McTadA were labeled at the top and bottom of the alignment.

Lastly, to check our hypothesis on the weak binding capacity of the enzyme, we performed binding studies using the ITC method. Along with it, TadA from EcTadA, BsTadA, SaTadA and Tad2/3 heterodimer from S. cerevisiae (ScTad2/3) were used as controls. To eliminate the interference of possible deamination during the binding process, we also mutated the key catalytic glutamate residue (Glu59 in EcTadA, Glu55 in BsTadA, Glu55 in SaTadA, respectively, and Glu56 in ScTad2). Most of these mutants except for BsTadA/E55Q turned out to be quite unstable and the proteins were also heavily bound by co-purified nucleic acids that were difficult to remove. As a result, they constantly precipitated from solution immediately upon elution from the affinity column (data not shown). Nevertheless, we managed to perform the ITC study using BsTadA/E55Q and WT McTadA. We found that while BsTadA showed quite strong affinity toward the Samini-RNA (2.8 ± 0.4 µM), McTadA did not show signs of a binding event (as suggested by the little amount of heat generated during the titration) even if we raised the protein and RNA concentrations to 35 µM and 300 µM, respectively (Figure 7). This result supported our theory that structural conflicts would arise upon the binding of tRNA and the inert activity was due to reduced binding affinity (KD) rather than catalysis (kcat). Additionally, EMSA was also carried out using full-length tRNAArg from M. capricolum or tRNAAla from S. cerevisiae. ScTad2/3 showed reasonable binding toward SctRNAAla as well as MctRNAArg (although with reduced affinity). In contrast, McTadA was not found to bind MctRNAArg at all (Supplementary Figure S2).

The ITC binding studies.

Figure 7.
The ITC binding studies.

The binding affinities of BsTadA/E55Q (A) and WT McTadA (B) to Samini-RNA were tested.

Figure 7.
The ITC binding studies.

The binding affinities of BsTadA/E55Q (A) and WT McTadA (B) to Samini-RNA were tested.

Discussion

The post-transcriptional modification scenarios present in tRNAs differ greatly from one organism to another, and represent the largest modification repertoire found in all RNA types. Nucleotide 34 in tRNA plays an essential role in determining the degenerate codons during the translation process by pairing with the last base of the corresponding codons. First proposed by Francis Crick more than a half century ago, the wobble hypothesis states that the third position of the codon on mRNA and the first position of the anticodon on tRNA could pair with each other in a less stringent manner than that of the canonical base pairs, and therefore expands the codon usages. In accordance with this theory, base 34, or the ‘so-called' wobble base is usually extensively modified to adapt to the different basepairing needs.

One frequently occurring modification is that the wobble A34 in some precursor tRNAs is enzymatically converted to inosine by specific tRNA:A34 deaminases during tRNA maturation. The deaminated inosine is capable of reading the C, U and A-ending codons, but the basepairing with a G-ending codon is not allowed [9]. Thus the A-to-I deamination provides more possibilities and diversities of pairing combinations, and is adopted by many organisms including eukaryotes. For example, in most bacteria, the four arginine codons (CGN) are decoded by two tRNAs. One tRNA harboring a wobble inosine (tRNAArgICG) reads the CGU, CGC and CGA codons, while a second tRNA harboring a wobble cytidine (tRNAArgCCG) reads the remaining CGG codon. However, M. capricolum and some closely related Mollicutes present exceptions to the rule, in that the gene encoding tRNAArgCCG or even tadA is missing. M. capricolum harbors only one copy of tadA and tRNAArgACG genes. As A34 deamination is considered an early event in tRNA maturation process, the 6% non-deaminated A34-tRNA molecules are not to be easily overlooked, comparing to the 100% A-to-I conversion observed in B. sublitis. On the other hand, decoding of the CGG codon still poses a problem, as I34-tRNAArgCCG would not be a good decoder. After studying the tRNAArg sequences of M. capricolum, Yokobori et al. [15] therefore deduced that tRNAArgACG molecules with a non-deaminated wobble adenosine in this organism is used to decode the CGG codon. To explain the poor activity toward McTadA, they proposed that the enzyme would bind poorly to its tRNA substrates, based on the homology modeling of the SaTadA-Samini-RNA complex.

To validate this interesting hypothesis, we solved the structures of the WT and a mutant form of McTadA in this study. Our crystallographic studies and structural analyses offered support to Yokobori's hypothesis in that multiple structural clashes against the RNA substrates would lead to a reduced association. In our in vitro activity assays, we did not observe any deamination activity from the McTadA in the recombinant form for unknown reasons. It is possible that the deamination process requires some pre-requisite tRNA modifications, which are not present on tRNA transcript. Additionally, we could not rule out the possibility that deamination catalyzed by McTadA may require an accessory protein or cofactor in vivo yet to be identified.

Upon structural comparison to that of EcTadA or SaTadA, we introduced site-specific mutations to the enzyme and tried to restore its activity. However, the TM mutant did not regain the deamination activity, and the expression failure of the quintuple mutant designed in the second-round mutagenesis prevented us from further engineering trials. Possible reasons for this result could come from the dimer interface differences. One should note that all our modeling efforts were merely based on the Samini-RNA substrate from the SaTadA complex (PDB 2B3J) [10], due to a lack of additional structural information on (t)RNA-bound form of TadAs. McTadA may adopt a distinct substrate binding mode, and in this case it calls upon the structure determination of McTadA in complex with its genuine RNA substrates.

The structural and sequence differences between McTadA and its orthologs are interesting. Among the three conflicts, the largest structural conflict comes from the Ser105-Ile110 fragment, in which Phe108 makes hydrophobic contacts with the terminal helix. These non-polar interactions fix this bulky and conserved phenylalanine residue in place, most likely to exclude the binding of RNAs as well as the base flipping of A34 and C35 (compare glycine or alanine in equivalent position in bacteria, Figure 6D). These two nucleosides are the most important recognition elements for the deaminase, especially for the yeast ScTad2/3 enzyme. Similarly, Asn22 is sterically incompatible with C35 and in many Mollicutes, and it is semi-conserved or replaced by iso-steric or larger residues in Mollicutes (glutamate or aspartate, Figure 6D) but is substituted in bacteria. In contrast, glycine is found in this position in most other bacteria with a fully functional TadA enzyme. In addition to the visible hindrance from the complex model, the enzyme displays a wider dimer interface, which also weakens the association of RNA. All known bacterial TadA enzymes work as homodimers, and only one subunit carries out the catalytic function whereas the other subunit only provides a fraction of support to RNA substrates. On the other hand, eukaryotic Tads work as heterodimers, but the Tad3 subunit is also a helper subunit due to the mutation of the key catalytic glutamate residue. As the two subunits of heterodimeric Tad each clearly have designated functions, the catalysis could not be achieved by a single subunit. In comparison, the second monomer of the McTadA contributes very little to the binding of RNA due to its loose association to the catalytic monomer, and it also creates structural conflicts around the Ser69 residue. Consequently, by reducing the possible interactions and increasing destructive hindrance with RNA substrates, both the binding and catalytic efficiency of the McTadA deamination reaction would be quite low. In a sense, McTadA ingeniously employs the dimer assembly to impair its own deamination function. Consequently, M. capricolum retains a portion of non-deaminated tRNAArgACG, which would make sure that all four arginine CGN codons are decoded.

The A34-tRNA deamination facilitates the codon recognition process and allows codon expansion according to the wobble hypothesis. This beneficial modification is widely present in both bacteria and eukarya. Consequently, Tad is an essential enzyme and the cells without this gene are not viable. However, M. capricolum and some Mollicutes choose to ‘sabotage' this modification event, since by retaining the A34-containing tRNA, they are equipped with the capability to decode yet an extra codon. The unmodified adenosine 34 in tRNAArg was reported in as early as 1984 [29]. In this work, Sibter and co-workers identified a yeast mitochondria tRNAArg belonging to the four CGN arginine codon family that contained unmodified A34. They tried to explain the decoding problem of the CGN codons with the classical ‘two out of three’ hypothesis, as the other two base pairs are of G-C combination. Moreover, the chloroplasts of A. thaliana encode only tRNAArgACG and tRNAArgUCU to decode the arginine codon. However, the knockdown of the chloroplastic TadA expression was proven non-lethal to plants [30], similar to the M. capricolum scenario studied here. In Yokobori's work, some species were found to go even further than M. capricolum, by losing the tadA gene (the Groups II and III Mollicutes) and relying completely on the ACG anticodon with an unmodified A34 to read the arginine four-codon box. This decoding strategy could be considered as another type of ‘superwobbling’ proposed by Rogalski et al. [31] in which a single base at the wobble position is employed to decode all degenerate codons and allows the viability of a species with a reduced set of tRNAs. However, as Yokobori et al. [15] have proposed, changing the decoding strategy through cellular evolution necessitates a series of pre-arranged events, such as complex point mutations in the modification enzymes (probably also in tRNAs) etc. Whether the extra efforts to evolve the TadA enzyme and corresponding tRNA genes are worthwhile in terms of cellular resources (energy, metabolite availability etc.) is an interesting question and warrants further investigation. For example, a possible scenario is that M. capricolum could retain a gene to encode tRNAArgCCG to save going through the troubles of mutations. The current non-classical decoding strategy discovered by Yokobori et al. is a result of co-evolution of both the tRNA and tadA genes. In addition, the partial and complete ‘bypass' of the A34 deamination process provides an alternative solution to the predicament of organisms with reduced genomes (thus less tRNA species) and faster evolving speed like Mycoplasmas or Mollicutes. This strategy may not be unique among all the organisms, but similar approaches can be certainly adopted by others.

Competing Interests

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

Funding

Funding for this work was provided by the National Natural Science Foundation of China 31870782, the Natural Science Foundation of Guangdong province 2018A030313313 and 2017A030310190.

Author Contribution

W.X. conceived and designed the research; H.L., S.W. and D.R. performed the research; W.X. analyzed data and wrote the paper. All authors reviewed the manuscript.

Data Availability

The atomic co-ordinates and structure factors have been deposited in the RCSB Protein Data Bank with accession numbers 6L2L and 6L2M.

Acknowledgements

We thank the BL19U1 beamline at Shanghai Synchrotron Radiation Facility for assistance during diffraction data collection. Funding for this work was provided by the National Natural Science Foundation of China 31870782, the Natural Science Foundation of Guangdong province 2018A030313313 and 2017A030310190.

Abbreviations

     
  • AaTadA

    Aquifex aeolicus TadA

  •  
  • ADAR

    adenosine deaminase acting on RNA

  •  
  • AfTadA

    Agrobacterium fabrum TadA

  •  
  • CDA

    cytidine deaminase

  •  
  • E. coli

    Escherichia coli

  •  
  • EcEndoV

    E. coli Endonuclease V

  •  
  • EcTadA

    Escherichia coli TadA

  •  
  • McTadA

    M. capricolum TadA

  •  
  • McTadA-TM

    the McTadA triple mutant

  •  
  • SaTadA

    Staphylococcus aureus TadA

  •  
  • SeTadA

    Salmonella enterica TadA

  •  
  • Tad

    tRNA-adenosine deaminase

  •  
  • WT

    wild type

References

References
1
Schmeing
,
T.M.
,
Voorhees
,
R.M.
,
Kelley
,
A.C.
,
Gao
,
Y.G.
,
Murphy
,
F.V.
,
Weir
,
J.R.
et al (
2009
)
The crystal structure of the ribosome bound to EF-Tu and aminoacyl-tRNA
.
Science
326
,
688
694
2
Motorin
,
Y.
and
Helm
,
M.
(
2010
)
tRNA stabilization by modified nucleotides
.
Biochemistry
49
,
4934
4944
3
Hopper
,
A.K.
(
2013
)
Transfer RNA post-transcriptional processing, turnover, and subcellular dynamics in the yeast Saccharomyces cerevisiae
.
Genetics
194
,
43
67
4
Helm
,
M.
,
Brule
,
H.
,
Degoul
,
F.
,
Cepanec
,
C.
,
Leroux
,
J.P.
,
Giege
,
R.
et al (
1998
)
The presence of modified nucleotides is required for cloverleaf folding of a human mitochondrial tRNA
.
Nucleic Acids Res.
26
,
1636
1643
5
Zhang
,
Q.
,
Zhang
,
L.
,
Chen
,
D.
,
He
,
X.
,
Yao
,
S.
,
Zhang
,
Z.
et al (
2018
)
Deletion of Mtu1 (Trmu) in zebrafish revealed the essential role of tRNA modification in mitochondrial biogenesis and hearing function
.
Nucleic Acids Res.
46
,
10930
10945
6
Wu
,
Y.
,
Wei
,
F.Y.
,
Kawarada
,
L.
,
Suzuki
,
T.
,
Araki
,
K.
,
Komohara
,
Y.
et al (
2016
)
Mtu1-mediated thiouridine formation of mitochondrial tRNAs is required for mitochondrial translation and is involved in reversible infantile liver injury
.
PLoS Genet.
12
,
e1006355
7
Björk
,
G.R.
(
1995
) Biosynthesis and function of modified nucleosides. In
tRNA: Structure, Biosynthesis, and Function
(
Söll
,
D.
and
RajBhandary
,
U.
, eds), pp.
165
205
,
ASM Press
,
Washington, DC
8
Curran
,
J.F.
(
1998
) Modified nucleosides in translation. In
Modification and Editing of RNA
(
Grosjean
,
H.
and
Benne
,
R.
, eds), pp.
493
516
,
ASM Press
,
Washington, DC
9
Auxilien
,
S.
,
Crain
,
P.F.
,
Trewyn
,
R.W.
and
Grosjean
,
H.
(
1996
)
Mechanism, specificity and general properties of the yeast enzyme catalysing the formation of inosine 34 in the anticodon of transfer RNA
.
J. Mol. Biol.
262
,
437
458
10
Losey
,
H.C.
,
Ruthenburg
,
A.J.
and
Verdine
,
G.L.
(
2006
)
Crystal structure of Staphylococcus aureus tRNA adenosine deaminase TadA in complex with RNA
.
Nat. Struct. Mol. Biol.
13
,
153
159
11
Kuratani
,
M.
,
Ishii
,
R.
,
Bessho
,
Y.
,
Fukunaga
,
R.
,
Sengoku
,
T.
,
Shirouzu
,
M.
et al (
2005
)
Crystal structure of tRNA adenosine deaminase (TadA) from Aquifex aeolicus
.
J. Biol. Chem.
280
,
16002
16008
12
Gerber
,
A.P.
and
Keller
,
W.
(
1999
)
An adenosine deaminase that generates inosine at the wobble position of tRNAs
.
Science
286
,
1146
1149
13
Tsutsumi
,
S.
,
Sugiura
,
R.
,
Ma
,
Y.
,
Tokuoka
,
H.
,
Ohta
,
K.
,
Ohte
,
R.
et al (
2007
)
Wobble inosine tRNA modification is essential to cell cycle progression in G(1)/S and G(2)/M transitions in fission yeast
.
J. Biol. Chem.
282
,
33459
33465
14
Kim
,
J.
,
Malashkevich
,
V.
,
Roday
,
S.
,
Lisbin
,
M.
,
Schramm
,
V.L.
and
Almo
,
S.C.
(
2006
)
Structural and kinetic characterization of Escherichia coli TadA, the wobble-specific tRNA deaminase
.
Biochemistry
45
,
6407
6416
15
Yokobori
,
S.
,
Kitamura
,
A.
,
Grosjean
,
H.
and
Bessho
,
Y.
(
2013
)
Life without tRNAArg-adenosine deaminase TadA: evolutionary consequences of decoding the four CGN codons as arginine in Mycoplasmas and other Mollicutes
.
Nucleic Acids Res.
41
,
6531
6543
16
Otwinowski
,
Z.
and
Minor
,
W.
(
1997
)
Processing of X-ray diffraction data collected in oscillation mode
.
Methods Enzymol.
276
,
307
326
17
Elias
,
Y.
and
Huang
,
R.H.
(
2005
)
Biochemical and structural studies of A-to-I editing by tRNA:A34 deaminases at the wobble position of transfer RNA
.
Biochemistry
44
,
12057
12065
18
Emsley
,
P.
,
Lohkamp
,
B.
,
Scott
,
W.G.
and
Cowtan
,
K.
(
2010
)
Features and development of Coot
.
Acta Crystallogr. D Biol. Crystallogr.
66
,
486
501
19
Emsley
,
P.
and
Cowtan
,
K.
(
2004
)
Coot: model-building tools for molecular graphics
.
Acta Crystallogr. D Biol. Crystallogr.
60
,
2126
2132
20
Adams
,
P.D.
,
Afonine
,
P.V.
,
Bunkoczi
,
G.
,
Chen
,
V.B.
,
Davis
,
I.W.
,
Echols
,
N.
et al (
2010
)
PHENIX: a comprehensive Python-based system for macromolecular structure solution
.
Acta Crystallogr. D Biol. Crystallogr.
66
,
213
221
21
Afonine
,
P.V.
,
Grosse-Kunstleve
,
R.W.
,
Echols
,
N.
,
Headd
,
J.J.
,
Moriarty
,
N.W.
,
Mustyakimov
,
M.
et al (
2012
)
Towards automated crystallographic structure refinement with phenix.refine
.
Acta Crystallogr. D Biol. Crystallogr.
68
,
352
367
22
Chen
,
V.B.
,
Arendall
,
W.B.
,
Headd
,
J.J.
,
Keedy
,
D.A.
,
Immormino
,
R.M.
,
Kapral
,
G.J.
et al (
2010
)
Molprobity: all-atom structure validation for macromolecular crystallography
.
Acta Crystallogr. D Biol. Crystallogr.
66
,
12
21
23
Wang
,
C.
,
Guo
,
Y.
,
Tian
,
Q.
,
Jia
,
Q.
,
Gao
,
Y.
,
Zhang
,
Q.
et al (
2015
)
SerRS-tRNA(Sec) complex structures reveal mechanism of the first step in selenocysteine biosynthesis
.
Nucleic Acids Res.
43
,
10534
10545
24
Zhang
,
Z.
,
Jia
,
Q.
,
Zhou
,
C.
and
Xie
,
W.
(
2015
)
Crystal structure of E. coli endonuclease V, an essential enzyme for deamination repair
.
Sci. Rep.
5
,
12754
25
Guo
,
G.
,
Ding
,
Y.
and
Weiss
,
B.
(
1997
)
Nfi, the gene for endonuclease V in Escherichia coli K-12
.
J. Bacteriol.
179
,
310
316
26
Gates
,
F.T.
and
Linn
,
S.
(
1977
)
Endonuclease V of Escherichia coli
.
J. Biol. Chem.
252
,
1647
1653
PMID:
[PubMed]
27
Weiss
,
B.
(
2001
)
Endonuclease V of Escherichia coli prevents mutations from nitrosative deamination during nitrate/nitrite respiration
.
Mutat. Res.
461
,
301
309
28
Krissinel
,
E.
and
Henrick
,
K.
(
2007
)
Inference of macromolecular assemblies from crystalline state
.
J. Mol. Biol.
372
,
774
797
29
Sibler
,
A.P.
,
Dirheimer
,
G.
and
Martin
,
R.P.
(
1986
)
Codon reading patterns in Saccharomyces cerevisiae mitochondria based on sequences of mitochondrial tRNAs
.
FEBS Lett.
194
,
131
138
30
Delannoy
,
E.
,
Le Ret
,
M.
,
Faivre-Nitschke
,
E.
,
Estavillo
,
G.M.
,
Bergdoll
,
M.
,
Taylor
,
N.L.
et al (
2009
)
Arabidopsis tRNA adenosine deaminase arginine edits the wobble nucleotide of chloroplast tRNAArg(ACG) and is essential for efficient chloroplast translation
.
Plant Cell
21
,
2058
2071
31
Rogalski
,
M.
,
Karcher
,
D.
and
Bock
,
R.
(
2008
)
Superwobbling facilitates translation with reduced tRNA sets
.
Nat. Struct. Mol. Biol.
15
,
192
198

Supplementary data