Wybutosine biosynthesis: Structural and mechanistic overview

Abstract

Over the last 10 years, significant progress has been made in understanding the genetics, enzymology and structural components of the wybutosine (yW) biosynthetic pathway. These studies have played a key role in expanding our understanding of yW biosynthesis and have revealed unexpected evolutionary ties, which are presently being unraveled. The enzymes catalyzing the 5 steps of this pathway, from genetically encoded guanosine to wybutosine base, provide an ensemble of amazing reaction mechanisms that are to be discussed in this review article.

Keywords:

• tRNA modification
• S-adenosylmethionine
• Radical-SAM enzymes
• Wybutosine
• Fe-S clusters

Introduction

Post-transcriptional modifications are one of the processing events that result in functional RNA molecules. To date, more than 100 chemically distinct modifications are described and they are found across the 3 kingdoms of life.¹ Among all RNAs, tRNAs exhibit the largest number and the widest variety of modifications.¹ Some of them are generated by relatively simple biosynthetic reactions involving methylation, thiolation, pseudouridylation, or dihydrouridine formation, and are found in most tRNAs.² The Anticodon Stem Loop (ASL) region particularly displays more complex modifications, which are proposed to contribute to the stabilization of the codon-anticodon pair, thereby maintaining the translational fidelity with high efficiency.² The distinctive chemistry of the modified nucleosides located at the ASL, mainly at the wobble position 34 and purine at position 37, is crucial for the decoding process, which may explain why both positions present the largest variety of hyper-modifications found among all tRNAs.² The biosynthetic pathway leading to these hyper-modifications requires several enzymatic steps for which the enzymological data are especially rich, though challenging.^3,4

One of the most structurally complex tRNA modifications is the formation of the wybutosine base (yW) that contains a fluorescent tricyclic fused aromatic base derived from a genetically encoded guanosine residue (Fig. 1A).⁵ The yW and its derivatives are found at position 37 of eukaryotic and archaeal phenylalanine tRNA (tRNA^Phe) (Fig. 1B) and their biosynthetic pathways have been described by excellent articles in great detail (Fig. 1D).^6-8 In this review, we intend to describe the 5 chemical reactions that generate yW, from a mechanistic and structural perspective, and to highlight the unique chemical versatility of S-adenosylmethionine (SAM or AdoMet) in this pathway. In these 5 enzymatic reactions, SAM is utilized as a source of 3 different chemical entities. While Trm5, TYW3, and TYW4, all use SAM as a source of methyl group, TYW2 and TYW1 use SAM as a source of α-amino-α-carboxypropyl group and 5′-deoxyadenosine radical (Fig. 1D), respectively.

Figure 1. (A) Chemical structure of wybutosine (yW), all appended groups are colored. (B) Secondary structure of yeast tRNA^Phe. (C) Chemical structure of SAM. (D) Biosynthetic pathway of yW. Trm5 methylates G₃₇ to produce m¹G₃₇; TYW1, a Radical-SAM enzyme, catalyzes the formation of 4-demethylwyosine (imG-14) using pyruvate as a co-substrate; TYW2 appends the α-amino-α-carboxypropyl moiety of SAM to produce 7-aminocarboxypropyl-demethylwyosine (yW-72); TYW3 methylates N⁴ of yW-72 to produce 7-aminocarboxypropyl-wyosine (yW-58); and TYW4 both transfers a carboxymethyl group and methylates yW-58 to yield yW.

The S-Adenosylmethionine (SAM)-Dependent tRNA-Methyltransferases Trm5, TYW3, and TYW4

The overwhelming majority of tRNA-methyltransferases use SAM as a methyl group donor.^9-11 SAM is an important biological sulfonium compound, which is involved in many essential biochemical processes.¹² It is one of the most widely used enzyme cofactors, second only to ATP.¹³ SAM is biosynthesized during the reaction of methionine with ATP, which is catalyzed by SAM synthetase or methionine adenosyltransferase.¹⁴ This reaction is stereospecific and only generates an S-configuration at the sulfur atom (Fig. 1C). The strong preference for SAM over other methyl donors, such as folate, reflects the highly favorable thermodynamics of SAM-dependent methyl-transfer reactions. The methylation of nucleophiles is the consequence of an intrinsic chemical property of SAM, namely the strong electrophilic character of its methyl group. This property has been exploited by SAM-dependent tRNA-methyltransferases, which bring SAM into contact with nucleophilic groups of substrates.

tRNA-methyltransferase Trm5

Trm5 is specific to eukaryotes and archaea, and is evolutionary unrelated to the bacterial counterpart TrmD.^15,16 Trm5 is a member of class I amino-methyltransferase family, which transfers the methyl group from SAM to nitrogen in widely different chemical environments.¹⁷ The crystal structures of the enzyme from the archaeon Methanocaldococcus jannaschii in complex with its substrate tRNAs and SAM have been reported.^18,19 The overall structure consists of 3 domains (D1, D2 and D3) in which D1 is linked to D2 by a flexible loop comprising 8 residues (Glu66-Lys73), whereas D2 and D3 are tightly associated (Fig. 2A). The arrangement of the 3 domains therefore suggests that D1 acts as the tRNA recognition entity similar to what has been reported for tRNA modification enzyme MiaA.²⁰ Unlike Trm5, MiaA is composed of 2 domains: the N-domain which is linked by a flexible loop to the C-terminal catalytic domain. Similar to N-domain of MiaA, D1 of Trm5 undergoes a drastic conformational change from apo to tRNA-bound structures (Fig. 2A & B). While D1 recognizes both variable and T arms, forming the outer corner of L-shape tRNAs, the catalytic domains D2 and D3 both directly interact with the anticodon region comprising the anticodon stem-loop and the D arm of both tRNA^Cys_GCA and tRNA^Cys_GCA in the reported structures.¹⁸ Interestingly, the structural analysis suggests that Trm5 has evolved to select and recognize L-shaped tRNAs as its substrates, and it thus functions as a tRNA tertiary structure checkpoint in the tRNA methylation process.¹⁸ It is likely that Trm5 enzyme enhances reaction rates by simply promoting favorable orientation and spatial proximity of the tRNA substrate and methyl group of SAM. Recent considerations from the kinetic and site-directed mutagenesis data in light of the Trm5-tRNA-SAM ternary structures suggest that the conserved R145 is involved in stabilizing a proposed oxyanion at G₃₇-O⁶ (Fig. 3), and E185 has been proposed to act as a general base to accept a proton from N¹-G₃₇, even though the side chain of E185 is approximately 12 Å away from N¹-G₃₇ (Fig. 2C).²¹

Figure 2. Crystal structure of archaeal Trm5. (A) Structure of Trm5 in complex with SAM (PDB id : 2YX1). The three domains of Trm5 are color coded and labeled. SAM is depicted as a stick model with black for carbon atoms. (B) Ternary structure of Trm5 in complex with tRNA^Leu and SAM (PDB id : 2ZZM). G₃₇ of tRNA^Leu is shown as a stick model with orange for carbon atoms. (C) The active site of Trm5 in which the side chains of R145 and the proposed catalytic residue E185 are shown as stick models. Figures 1–6 are made using PyMOL (www. pymol. org).

Figure 3. Proposed catalytic reaction mechanism for Trm5. R: (CH₂)₂-CH(NH₂)COO⁻.

tRNA-methyltransferase TYW3

The fourth step in the yW biosynthesis relies on TYW3, an S-AdoMet-dependent enzyme, which methylates the N³ position of the imidazo-purine ring to yield yW-58 (Fig. 1D). The gene encoding this enzyme (tyw3) has first been identified by ribonucleome screening method in S. cerevisiae.⁶ Later, archaeal TYW3 orthologues were identified using S. cerevisiae and Arabidopsis thaliana as representatives of eukaryotes.⁸ TYW3 from different organisms was found to be a highly conserved protein. However, homologues of TYW3 from A. thaliana and Oryza sativa have been shown to contain an extended C-terminal domain of about 1000 amino acid residues. Interestingly, this extension is almost identical to the C-terminal domain of TYW4 and the entire sequence of TYW2. Thus, the plant enzymes appear to be a large fusion protein comprising TYW3, the C-terminal domain of TYW4, and the entire TYW2.

Curiously, the fourth step in yW biosynthesis is the only one in the pathway that escaped biochemical and enzymological investigations. However, 5 highly homologous crystal structures from Archaeoglobus fulgidus (PDB id: 2QG3), Sulfolobus solfataricus (PDB id: 1TLJ), Pyrococcus horikoshii (PDB ids: 2DRV, 2IT3, and 2IT2), have recently been deposited in Protein Data Bank (Fig. 4A). According to PFAM (http://pid. nci. nih. gov/2011/110913/full/pid. 2011. Three. shtml), all these apo structures are annotated as TYW3-like methyltransferases. The crystal structures show that all of these proteins form head-to-head dimers in each crystal lattice. Whether the biological unit of these proteins is monomer or dimer remains to be investigated, and in the absence of a SAM-bound structure, we can only speculate the probable SAM binding pocket and the active site of these proteins (Fig. 4B). As shown in Figure 4, the representative protein is composed of 2 domains, and the C-terminal domain contains a long C-terminal helix that traverses the entire length of 2 domains (Fig. 4A). There is a cavity at the interface of 2 domains (Fig. 4A & B) where several invariant residues reside (Fig. 4B & C). The strictly conserved nature of these residues (Fig. 4C) suggests that they probably play important roles in catalysis. From a mechanistic point of view, the methylation of N³ atom is reminiscent of the reaction catalyzed by Trm5. We therefore anticipate that TYW3 enzyme uses a similar mechanism for N³ methylation, which generates yW-58 (Fig. 3).

Figure 4. Crystal structure of yeast TYW4. (A) Apo structure of TYW4 (PDB id : 2ZWA). The three domains of TYW4 are colored and labeled. The N-terminal helices are shown as yellow. (B) Structure of TYW4 in complex with SAM (PDB id : 2ZW9). SAM is depicted as a stick model with black for carbon atoms. The N-terminal helix is shown as yellow. (C) The active site of TYW4 in which the side chains of Y229 and R88, and SAM are shown as stick models. The flexible loop of C-domain is shown in magenta and is labeled. The substrate binding site is labeled.

tRNA-methyl-methoxycarbonyltransferase TYW4

TYW4 catalyzes the final step in yW biosynthesis, i.e. the modification of the α-amino-α-carboxypropyl side chain attached by TYW2 to the C⁷ position of the tricyclic ring (see below) (Fig. 1). The gene encoding this enzyme (tyw4) has been identified by ribonucleome screening analysis in S. cerevisiae.⁶ An in vitro biochemical analysis coupled to mass spectrometry showed that TYW4 is a bi-functional enzyme that catalyzes 2 reactions (i.e., methylation and methoxycarbonylation) of the α-amino-α-carboxypropyl side chain apparently within the same catalytic site.²² Interestingly, the methoxycarbonylation reaction proceeds through the fixation of CO₂, an oddity among all RNA modification mechanisms.

The apo and SAM-bound TYW4 structures have been obtained to resolution 2.5 Å and 2.7 Å, respectively. The overall structure of the enzyme consists of an N-terminal domain (1–350) and a C-terminal domain (371–694), connected by a linker (residues 351–370) (Fig. 5A). The structures reveal that the N-terminal domain belongs to the class I SAM-dependent methyltransferases, while the C-terminal domain adopts a β-propeller fold, which has a flexible loop that resides near the yW-58 binding pocket. Upon SAM binding, the N-terminal α-helix undergoes a significant conformational change so as to efficiently shield the active site from the solvent (Fig. 5A & B).²²

Figure 5. Crystal structure of a putative TYW3 from Archaeoglobus fulgidus. (A) Apo structure of TYW3 (PDB id : 2QG3). The secondary structural elements, β-strands, α-helices, and loops, are colored as cyan, yellow, and light-magenta. The C-terminal α-helix is colored as dark blue. The side chains of invariant residues are shown with stick models. (B) Electrostatic surface potential of TYW3. Blue, red, and white patches present the positively, negatively-charged, and neutral regions of the protein surface, respectively. The presumed SAM binding pocket and the active site are labeled. (C) The presumed SAM-binding and the active sites of TYW3. The side chains of strictly conserved residues are shown as stick models.

In the absence of a ternary structure of TYW4 in complex with SAM and a tRNA substrate, docking exercises for such a complex were performed with yW-58 as a substrate. In the resulting model, it was proposed that the α-amino-α-carboxypropyl of yW-58 adopts 2 different binding modes, which are responsible for the 2 reactions described above. For the methylation reaction, TYW4 is supposed to enhance the reaction rate by simply promoting favorable orientation and spatial proximity of the inherently nucleophilic oxygen atom of the α-carboxyl group toward the electrophilic methyl group of SAM. The amino acid residues Arg88 and Tyr229 have been proposed for this improved positioning (Fig. 5C and Fig. 6A). In contrast, for the methoxycarbonylation reaction, the hydroxyl group of Tyr229 is proposed to act as a general base catalyst to deprotonate the α-amino group of yW-58, while Arg88 would activate CO₂ making it prone to be methylated by SAM (Fig. 6B).

Figure 6. Proposed catalytic reaction mechanism for both methylation (A) and carboxymethylation (B) for TYW4.

tRNA-aminocarboxypropyltransferase TYW2

TYW2 catalyzes the transfer of the 3-amino-3-carboxypropyl group (acp) from SAM to the C⁷ position of the tricyclic core structure of imG-14 base, forming yW-72 (yW minus 72 Da) (Fig. 1C).^6,23 Reactions in which acp-group is derived from SAM for substrate modification have also been reported for several biosynthetic pathways including diphthamide,²⁴ the antibiotic nocardicin,²⁵ diacylglyceryl-O-4′-(N, N, N, -trimethyl) homoserine,²⁶ and acp³U modified nucleoside in tRNAs.²⁷ Reaction mechanisms have yet to be delineated for the latter 3 biosyntheses. In the case of wybutosine, the detailed mechanism of acp-group transfer catalyzed by TYW2 is also poorly understood. However, it has been suggested that the acp-group transfer proceeds via an ionic mechanism based on crystal structures of the enzyme in complex with SAM and 5′-deoxy-5′-methylthioadenosine (MeSAdo) and by mutagenesis studies²³ (Fig. 7A & B). Thus, TYW2 likely transfers the acp-group through a single-displacement mechanism in which the inversion of the configuration of the methylene carbon occurs during the nucleophilic attack by C⁷ in the imG-14 base. Generally, nucleophilic attacks by aromatic carbon atoms are more demanding than those by polarized nitrogen and oxygen atoms. For example, it is well known that the SAM-dependent C⁵ methylation of pyrimidine by E. coli Fmu methyltransferase involves activation of the nucleo-base through covalent-bond formation between a conserved Cys-thiol of the enzyme and the C⁶ carbon of the pyrimidine substrate.²⁸ This covalent-bond formation increases the nucleophilicity of the C⁵ carbon and thus facilitates its alkylation. Although TYW2 catalyzes the alkylation of aromatic carbon atom, the published structure showed no conserved Cys residues near the catalytic site.²³ It was therefore suggested that the nucleophilicity of the C⁷ carbon atom might be increased via wyosine tautomerization.²⁹

Figure 7. Crystal structure of P. horikoshii TYW2. (A) SAM-complexed structure of TYW2 (PDB id : 3A25). The two domains of TYW2 are color coded and labeled. SAM and the substrate binding pocket are labeled. (B) The active site of TYW2 in which the side chains of M107, N112, R116, N112, and SAM are shown as stick models. SAM and the location of the substrate binding pocket are labeled.

A completely different mechanism for C-C bond on the aromatic ring has recently been unveiled during the investigation of diphtamide biosynthesis in the archaeon Pyrococcus horikoshii strain.²⁴ Indeed, structural and biochemical data showed that Dph2, the enzyme involved in the acp-group transfer from SAM to the histidine residue in the translation elongator factor 2 (eEF2) is a SAM-dependent, [4Fe-4S]-containing enzyme.²⁴ The reduced [4Fe–4S]⁺ cluster provides one electron to reductively cleave the C_{γ, Met}–S bond of SAM and generate a 3-amino-3-carboxypropyl radical along with methylthioadenosine.²⁴ The acp radical was then suggested to attack the histidine residue in the translation elongation factor 2 (eEF2).²⁴ Despite the fact that acp-group transfers from SAM are very similar in both Dph2 and TYW2, it is intriguing that these enzymes use completely different strategies to achieve alkylation of an aromatic ring.

tRNA-4-demethylwyosine synthase TYW1

The second step in yW biosynthesis is chemically the most challenging reaction: the imidazoline ring of yW is built on the m¹G₃₇ backbone of tRNA. This suggests that in addition to the tRNA substrate and SAM co-substrate, a second co-substrate, which is a source of the inserted 2-carbon unit, is also involved in the reaction.⁶ TYW1, the enzyme that catalyzes the transformation of m¹G₃₇ into ImG-14 has previously been classified as a Radical-SAM enzyme due to the presence of a conserved CysxxxCysxxCys motif in its aminoacid sequence⁶ In 2007, 2 apo structures of TYW1 from Methanococcus jannaschii and Pyrococcus horikoshii have been reported^30,31 (Fig. 8A & B). At that time, no in vitro assay was available, as the identity of the second co-substrate was unknown. In 2011, with the discovery that pyruvate is the 2-carbon donor for the reaction, the investigation of the mechanism was made possible.³² More recently, important insights into the mechanisms of TYW1 have been obtained.³³ First, reconstituted TYW1 was shown to contain 2 oxygen-sensitive [4Fe−4S]^2+/+ clusters, each ligated by only 3 cysteine residues that are absolutely required for activity.³³ Second, the N-terminal [4Fe-4S] cluster (cluster II), bound to the polypeptide by a CysX₁₂CysX₁₂Cys motif, contains a non-cysteinyl iron that is capable to bind and activate the pyruvate co-substrate.³³ Third, using a tRNA^Phe transcript from S. cerevisiae, which was labeled with (¹³C²H₃)-methyl-SAM on m¹G in a Trm5-dependent reaction, m¹G was shown to be the true substrate of the initial Ado• reaction that leads to the formation of imG-14 (T. Molle et al., to be published). These findings place substantial constraints on the mechanism by which the imidazoline ring is formed, as depicted by a working model in Scheme 4.

Figure 8. Crystal structures of TYW1. (A) Apo structure of archaeal TYW1 (PDB id : 2Z2U). The side chains of invariant cysteine residues are depicted as stick models and are labeled. (B) Apo structure of P. horikoshii TYW1 (PDB id : 2YX0). The side chains of invariant cysteine residues are depicted as stick models and are labeled.

Figure 9. Proposed catalytic reaction mechanism of holo-TYW1. Radical-SAM cluster I and cluster II are sketched as cubes, non-cysteinyl irons are shown in blue (cluster I) and red (cluster II), (A): SAM and pyruvate binding to cluster I and II respectively (reaction 1). Reduction of cluster I (reaction 2). Reductive cleavage of SAM in presence of m¹G₃₇-tRNA.

In this model, the reaction starts with the binding of SAM to cluster I and pyruvate to cluster II (Scheme 4A, reaction 1). In presence of electrons, cluster I-SAM complex is reduced and cluster II-pyruvate complex remains oxidized (Scheme 4A, reaction 2). When the substrate m¹G₃₇-tRNAs is added to the reaction mixture, the reductive cleavage of SAM occurs, which generates the canonical Ado• radical (Fig. 9A reaction 3) with concomitant release of methionine. In the following step, Ado• initiates H-atom abstraction from the methyl group of m¹G₃₇-tRNA (Fig. 9B, reaction 4), which results in formation of the substrate radical, while pyruvate remains bound to oxidized cluster II. Addition of this substrate radical to the C₂ of pyruvate initiates the homolytic cleavage of the C₁–C₂ bond and release of one electron and CO₂ (Fig. 9B, reaction 5). The last 2 steps of the mechanism consist in the formation of the tricyclic ring system through nucleophilic attack of the nitrogen atom on the C₃ of the pyruvate and elimination of H₂O (Fig. 9B, reactions 6 and 7). Alternatively, reactions 5 and 6 of Figure 9B could be reversed, that is, the initial nucleophilic attack of the exocyclic G₃₇ amino group on the cluster II-activated carbonyl is followed by radical coupling. A variant of this mechanism has been proposed in which pyruvate initially forms a Schiff base with a conserved and essential lysine (K41) upstream of the Radical-SAM cluster.^32,34 However, no evidence for such an adduct could be obtained either by UV-vis spectroscopy or by reductive trapping as commonly used for characterization of Schiff bases in PLP-containing enzymes.³⁵ On the contrary, preliminary Electron Paramagnetic Resonance (EPR) data obtained with the K41A mutant suggest that K41 is not at bonding distances of cluster II and rather plays an essential role in controlling the redox state of cluster II (T. Molle et al., to be published).

Several questions remain unsolved in this model, which are the focus of ongoing studies in our group. Two pressing questions to be addressed are: 1) What is the molecular basis of the interaction between cluster II and pyruvate; 2) How does cluster II control its redox potential. In this regard, parallel studies of another group of Radical-SAM enzymes involving radical insertion such as methylthiotransferases may help in deciphering the complex mechanism at work in TYW1.

Additional Pathways

Archaeal specific modifications : Involvement of aTrm5a and TYW3 enzymes

One recent report gives additional insights into the large variety of wyosine derivatives found in Archae.³⁶ Three families of methyltransferases that are homologous to Trm5 have been identified (aTrm5a-c) of which aTrm5a has the ability to methylate not only the genuine coded guanine, as found for Trm5, but also the imG-14 wyosine intermediate so as to give rise to isowyosine imG2 (7-methyl, imG-14). A classical methylation of the latter by TYW3 brings about 7-methylwyosine mimG (4,7-dimethyl, imG-14). How these enzymes mediate the transfer of a methyl group either to a nitrogen atom (Trm5) or a carbon (aTrm5a) awaits further biochemical and structural characterization.

tRNA-hydroxylase enzyme TYW5

The hydroxywybutosine (OHyW), a hydroxyl derivative of yW, is one of the additional modifications found in eukaryotes including human. It has been reported that JumonjC-related enzyme (JmjC) catalyzes the hydroxylation of yW to hydroxywybutosine (OHyW).³⁷ The JmjC-domain-containing protein, TYW5 (tRNA-yW-synthesising enzyme 5), catalyzes the formation of OHyW nucleoside, using Fe(II) ion and 2-oxoglutarate as cofactors. The crystal structure of human TYW5(hTYW5), alone and in complex with 2-oxoglutarate and Ni(II) ion, as an inactive substitute for Fe(II), have recently been reported to resolution 2.5 Å and 2.8 Å respectively ³⁸ (Fig. 11). The structures reveal that the catalytic domain consists of a β-jellyroll fold, a hallmark of the JmjC domain and other Fe(II)/2-oxoglutarateoxygenases.³⁸ The active site is located at the center of the β-jellyroll fold and consists of a highly conserved His-X-(Asp/Glu)-X_n-His signature motif.³⁹ These conserved residues provide 3 chemical groups to chelate the Fe(II) ion, while 2 additional chelating groups are supplied by the C₁ carboxylate and C₂ ketone groups of 2-oxoglutarate cofactor. In all these enzymes, the oxidative decomposition of 2-oxoglutarate to CO₂ and succinate leads to formation of a highly active Fe(IV)-oxo species that has a great potential for substrate hydroxylation (Scheme 5).

Figure 10. Proposed catalytic reaction mechanism for hTYW5.

Figure 11. Crystal structures of human TYW5. (A) Structure of hTYW5 in complex with Ni²⁺ and 2-oxoglutarate (PDB id : 3AL6). The two domains are depicted with different colors. Ni ion is shown as a dark green sphere, whereas 2-oxoglutarate (2-OG) and side chains of residues in the active site are shown with stick models. (B) The active site of TYW5. The side chains of invariant residues in the active site, Ni ion, and 2-OG are depicted as stick models and are labeled.

Conclusion

In 2004, we published a review article highlighting the amazing biological versatilities of SAM contrasting with its main known chemical use as a methylating reactant.¹² Indeed, in the cellular context, SAM is used as a source of methylene groups (in the synthesis of cyclopropyl fatty acids,⁴⁰ and rRNA modification⁴¹), amino groups (in the synthesis of 7, 8-diaminoperlagonic acid, a precursor of biotin⁴²), ribosyl groups (in the synthesis of epoxyqueuosine, a modified nucleoside in tRNAs^43,44), aminopropyl groups (in the synthesis of ethylene and polyamines⁴⁵), and 3-amino-3-carboxypropyl group (in the synthesis of acp³U, a modified nucleoside in tRNA ⁴⁶ and in the biosynthetic pathway of diphthamide ²⁴). It is remarkable that this unique situation is mirrored in one biosynthetic pathway, that of yW, which is described in the present review article. Indeed, in the 5 enzymatic reactions leading to yW, SAM is utilized as a source, of methyl group by Trm5, TYW3, and TYW4 (Fig. 1C), of α-amino-α-carboxypropyl group by TYW2 (Fig. 1C), and of 5′-deoxyadenosine radical by TYW1 that critically plays in initiating complex radical-mediated transformations (Fig. 1C). Recent reports on Radical-SAM enzymes suggest further unexpected usage of SAM, for example, as a carbide donor in maturation of nitrogenase.⁴⁷

As shown in this review, the mechanistic and structural enzymology of tRNA modifications can be considered a well investigated domain, as the research in this field has been occurring at an unprecedented rate. Recent published work on complex biosynthetic pathways leading to queuosine,⁴⁸ methylthioadenosine ⁴⁹ in tRNA and methyladenosine in rRNA ⁴¹ suggest that we are only at the beginning of an exciting story.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Funding

This work was supported by ANR Blanc-2010 grant INSERAD to PP-L, and Région Rhône-Alpes grants CIBLE 2011–2014 to TM. NIH grant 2U54GM75026 to the Northeast Structural Genomics Consortium.