U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Madame Curie Bioscience Database [Internet]. Austin (TX): Landes Bioscience; 2000-2013.

Cover of Madame Curie Bioscience Database

Madame Curie Bioscience Database [Internet].

Show details

Class I Lysyl-tRNA Synthetases

, , , , and .


Lysyl-tRNA synthetases are unique amongst the aminoacyl-tRNA synthetases in that they are found as both class I and class II enzymes. Most bacteria and all eukaryotes contain a class II LysRS whereas most archaea and a few bacteria contain a less common class I LysRS. In bacteria the class I LysRS is only found in the α-proteobacteria and a scattering of other groups including the spirochetes, while the class I protein is by far the most common form of LysRS in archaea. Functional and structural characterizations have shown that class I and II LysRS proteins are functionally equivalent but structurally unrelated. Consequently, despite their lack of sequence similarity, the class I and II LysRSs are able to recognize the same amino acid and tRNA substrates providing an example of functional convergence by divergent enzymes.

Distribution of the Class I Type LysRS

The 20 aminoacyl-tRNA synthetase (aaRS) proteins, as for example found in Escherichia coli,1 are divided into two mutually exclusive structural groups of ten members each termed class I and class II.2-4 The assignment of an aaRS specific for a particular amino acid to one of the two structural classes is almost completely conserved in the living kingdom, reflecting the ancient evolutionary origin of this family of proteins.5

The twenty canonical aaRS proteins do not represent the sole means of cellular aa-tRNA synthesis. It was first observed in Bacillus megaterium that Gln-tRNAGln could be synthesized via Glu-tRNAGln without the use of glutaminyl-tRNA synthetase (GlnRS).6 Subsequent studies showed that comparable indirect pathways exist for the production of Asn-tRNAAsn, which like Gln-tRNAGln is made by a tRNA-dependent amidotransferase, and selenocysteinyl-tRNASec (see Chapters 28 and 29 for reviews). While the use of these pathways was initially assumed to be idiosyncratic to a small minority of organisms, recent analyses of completed microbial genome sequences have shown them to be widespread in prokaryotes with the corresponding direct pathways absent.7 GlnRS is absent from all characterized archaea and the majority of bacteria, while asparaginyl-tRNA synthetase (AsnRS) is absent from most archaea and a significant number of bacteria.8 Initial analyses of the complete genome sequences of the archaea Methanococcus jannaschii 9 and Methanobacterium thermoautotrophicum10 suggested that in addition to AsnRS and GlnRS, genes encoding the cysteinyl-tRNA (CysRS) and lysyl-tRNA (LysRS) synthetases might also be absent from some organisms. Experimental studies showed that while both CysRS11-13 and LysRS14 are present in archaea, they are encoded by novel or reassigned genes making the original annotation of the corresponding open reading frames difficult. In the case of LysRS experimental approaches identified a gene encoding a class I aaRS (lysK) in the archaeon Methanococcus maripaludis, in contrast to all previously characterized LysRS proteins that belonged to class II. Comparative genomic analyses showed class II LysRS-encoding genes (lysS) to be absent from most, but not all, archaea and some bacteria, with lysK being found instead in all cases.15 Primarily as a result of the sequencing of numerous microbial genomes, over 50 lysK gene sequences are now known. The distribution of organisms containing lysK encompasses most of the archaea, a scattering of bacteria, but no eukaryotes. In nearly all cases lysK is found only when lysS is absent, the only exception found to date being the archaeal Methanosarcinaceae group where both are present. The class I and class II LysRSs function together in the Methanosarcinaceae to aminoacylate the specialized tRNAPyl(CUA) suppressor species.16 Lysyl-tRNACUA has been proposed to be a precursor in an as yet undefined pathway for the co-translational insertion of pyrrolysine at specific in frame amber codons.17,18

Biochemistry of the Class I LysRS

General Features

Functional19 and structural20 characterizations have shown that class I and II LysRS proteins are functionally equivalent but structurally unrelated. Consequently, despite their lack of sequence similarity, the class I and II LysRSs are able to recognize the same amino acid and tRNA substrates both in vitro and in vivo, providing an example of functional convergence by divergent enzymes.21 The two classes of LysRS proteins approach their RNA substrates from opposite sides, but recognize the same regions of tRNALys, namely the anticodon, acceptor stem and discriminator base.19,20 Within these common recognition sites in tRNALys the relative importance of particular nucleotides varies for the two classes of LysRS.21 This demonstrates how the unrelated forms of LysRS perform the same cellular function, in this case tRNALys recognition, using different molecular mechanisms. Mechanistic differences have also been observed for the initial steps in lysine recognition. The class II LysRS initiates aminoacyl-tRNA synthesis using only lysine and ATP to generate an enzyme bound aminoacyl-adenylate, as do all class II and the majority of class I aaRSs, whereas the class I LysRS requires tRNALys binding prior to aminoacyl-adenylate synthesis, a feature shared by only the class I aaRSs GlnRS, glutamyl- and arginyl-tRNA synthetase.19

tRNALys Identity

In vitro studies of tRNALys variants have shown that the identities of the anticodon and discriminator bases, and to a lesser extent the acceptor stem base pairs, are important for optimal recognition by the class I LysRS.19,21 In addition, the G:U wobble pair found in the acceptor stem of B. burgdorferi tRNALys1 (fig. 1) serves as an anti-determinant for E. coli LysRS recognition. It was also shown that the modifications of bases U34 and A37, variations of which have been identified in numerous lysine tRNAs,22,23 are not required for tRNA recognition.19,21 This provides an interesting contrast to E. coli GluRS, which is dependent on the presence of the same modification of U34 for optimal tRNA recognition.24 Nevertheless, the finding that many archaeal in vitro transcribed lysine tRNAs are not substrates for the corresponding class I LysRS raises the possibility that other nucleotide modifications may be important for recognition.

Figure 1. A) Predicted secondary structure of unmodified tRNALys1 from B.

Figure 1

A) Predicted secondary structure of unmodified tRNALys1 from B. burgdorferi. The anticodon and discriminator bases are circled and the G2:U71 wobble pair is boxed. B) Secondary structure of wild-type E. coli tRNALys. D, dihydrouridine; mnmsU, 5-methylaminomethyl- 2-thiouridine; (more...)

More detailed examination of the recognition of partially modified E. coli tRNALys anticodon variants by B. burgdorferi and M. maripaludis class I LysRSs revealed differences in their patterns of anticodon recognition.21 U35 and U36 were both found to be important for recognition by the B. burgdorferi enzyme, whereas only U36 played a role in recognition by the M. maripaludis class I LysRS. Further characterization of anticodon recognition by other class I LysRSs suggested that predominant recognition of U36 is a general characteristic, with recognition of both U35 and U36 being confined to the phylogenetic grouping containing Borrelia and Pyrococcus species (fig. 2).25 Investigation of an archaeal class II LysRS (from Methanosarcina barkeri) revealed that this enzyme recognizes all three anticodon nucleotides in tRNALys with U35 the most important position.25 Almost identical results have previously been reported in both bacterial21,26 and eukaryotic27 class II LysRSs. These findings suggest that the class II LysRSs utilize an evolutionarily conserved mechanism of tRNA anticodon recognition that is the same in archaea, bacteria and eukaryotes. This is in sharp contrast to the class I LysRSs, which show considerable diversity in how they recognize the anticodon of tRNALys.

Figure 2. Maximum likelihood phylogenetic tree of class I LysRS sequences.

Figure 2

Maximum likelihood phylogenetic tree of class I LysRS sequences. The tree has been rooted using GluRS sequences. Numbers indicate the percentage occurrence of nodes after 10,000 puzzling steps. From reference 25.

LysRS1 Structure

Overall Structure

The class I LysRS of Pyrococcus horikoshii has been crystallized and the structure of this protein solved both alone (at 2.6 Å resolution) and in a complex with L-lysine (at 2.9 Å resolution).20 Class I LysRS exhibits striking geometrical similarity with Thermus thermophilus glutamyl-tRNA synthetase (GluRS) (fig. 3A,B), whose crystal structure has been solved alone28 and in complex with tRNAGlu.29 The structural similarities cover the Rossmann fold domain, the SC-fold domain, and the α-helical hemispheric domain at the C-terminus, confirming the prediction that the class I LysRS is a member of subclass Ib.4 The most characteristic resemblance is observed in the α-helical hemispheric domain; the corresponding domain of GluRS consists of six, in contrast to five for LysRS, α helices folding into a distinctive fold named the “α-helix cage”.28 The α-helix cage of P. horikoshii LysRS is only the second example of such a domain found in all protein structures determined thus far.

Figure 3. Crystal structures of E.

Figure 3

Crystal structures of E. coli LysRS-II (PDB code, 1E24) A) and P. horikoshii LysRS-I (PDB code, 1IRX) B). The enzyme-bound cognate amino acid substrates are shown by CPK models. Adapted from reference 20.

Structural Basis of Lysine Recognition

In the L-lysine•class I LysRS (LysRS-I) complex, an L-lysine molecule is bound to the pocket formed on the Rossmann fold domain. The position of the L-lysine on LysRS-I is the same as that of the L-glutamate on the T. thermophilus GluRS, when the two aaRS structures are superposed. The ε-amino group of L-lysine provides a hydrogen bond with the side-chain hydroxyl group of Tyr268, and an electrostatic interaction with the side-chain carboxyl group of Glu41 (fig. 4A). The long aliphatic side chain of L-lysine interacts hydrophobically with the aromatic side chains of Trp218 and His240 (fig. 4A). The α-carboxyl group provides a weak hydrogen bond with the γ-O group of Thr29, and is also recognized by a water-mediated hydrogen bond to Trp218 (fig. 4A). The α-amino group hydrogen-bonds to the main-chain carbonyl group of Gly27 (fig.4A). As compared with the ligand free structure, Glu41, Trp218, and His240 slightly change their conformations upon binding with L-lysine. Comparison of the recognition mode of L-lysine by LysRS-I with that by E. coli class II LysRS (LysRS-II)30 indicates that while the active site architectures are fundamentally different the strategies for L-lysine recognition show some similarity (fig.4A,B); in both cases the ε-amino group is recognized by Glu and Tyr residues, and the aliphatic side chain hydrophobically interacts with aromatic residues.

Figure 4. The L-lysine-binding pocket of P.

Figure 4

The L-lysine-binding pocket of P. horikoshii LysRS-I A) and E. coli LysRS-II B). The bound L-lysine molecules are indicated, with the line pointing to the ε-amino group. The distances in Å between two charged or polarized atoms are indicated. (more...)

tRNA Recognition

The high degree of overall structural homology between GluRS and LysRS-I was used to develop a docking model of P. horikoshii LysRS-I and tRNA, by superposing the α-helix cage domain (which is responsible for tRNA anticodon binding by GluRS) and the Rossmann fold (active site) domain of the LysRS-I independently onto the same domains of the GluRS-tRNA co-crystal structure (fig. 5). In this docking model, the surface of the LysRS-I α-helix cage domain fits snugly with the three anticodon nucleotides, as also seen in the T. thermophilus GluRS:tRNAGlu structure. 29 For GluRS, all three anticodon nucleotides (C34, U35, and C36) of tRNAGlu are recognition elements.31 In contrast, biochemical analysis of tRNALys recognition by P. horikoshii LysRS-I indicated that there is no base-specific recognition of U34. Instead U35 and U36 are recognized, with U35 being the more important of the two, a result compatible with the proposed docking model.20

Figure 5. The docking model of LysRS-I and tRNA.

Figure 5

The docking model of LysRS-I and tRNA. The docked tRNA molecule is shown as a tube. The bound L-lysine and Zinc atoms and a modeled ATP molecule are shown. The amino acid residues proposed to interact with the tRNA anticodon and the acceptor end are indicated. (more...)

In addition to the anticodon, the discriminator base (N73) and acceptor stem function in tRNALys recognition by LysRS-I.19 In the GluRS:tRNAGlu structure, Arg47 interacts with the discriminator base A73.29 This residue is conserved as Arg in LysRS-I in the structure-based sequence alignment with GluRS (not shown), and the corresponding Arg72 of P. horikoshii LysRS1 is in close proximity to the discriminator A73 in the docking model (fig. 5). In addition, the docking model suggests that Lys238 and Asp239 are close to the 5' strand of the tRNA acceptor stem, and Lys274 is in proximity to nucleotide 25 (fig. 5).

Evolution of Class I LysRS

Class I LysRS Molecular Phylogenies

The phylogenetic tree for class I LysRS sequences contains two major branches broadly corresponding to the bacteria and the archaea (fig. 2). The inclusion of GluRS sequences places the root of the tree between the bacterial and archaeal branches, indicating that the evolutionary pattern of class I LysRS sequences is in agreement with that of the universal phylogenetic tree derived from ribosomal RNA sequences.25 While the class I LysRS tree appears canonical it is somewhat restricted, lacking any known eukaryotic examples and being mainly limited to α-proteobacteria in the bacterial branch and euryarchaeota in the archaeal branch. This observation of a limited canonical phylogeny among the class I LysRSs suggests that selective retention of the lysK gene in certain lineages has contributed substantially to their present distribution, as previously suggested.32

Closer examination of the class I LysRS phylogeny, however, suggests that limited gene transfer has also contributed to the phylogenetic distribution of class I LysRSs. One possible exception to the canonical universal phylogeny among the class I LysRSs is seen for the C. symbiosum protein, which is positioned within the main bacterial branch. Earlier analyses based upon smaller data sets placed C. symbiosum as a branch preceding the α-proteobacteria,33,34 possibly indicating horizontal transfer from an ancestral crenarchaeote. A more reliable interpretation of the exact positioning of the C. symbiosum protein within the class I LysRS tree, and of possible gene transfer events, is dependent on the availability of other closely related archaeal sequences. Within the archaeal branch of the class I LysRS tree, the group containing Pyrococcus species also contains several bacterial examples of LysRS. Functional analyses have shown that bacterial (B. burgdorferi) and archaeal (P. horikoshii) enzymes from within this group share common tRNALys anticodon recognition properties (see above) not seen for class I LysRSs from elsewhere in the phylogenetic tree, supporting a possible common origin. The positioning of this group within the canonical archaeal branch suggests that the corresponding bacterial LysRSs (Borrelia, Treponema and Streptomyces) originated from a gene transfer event from the pyrococcal progenitor into an ancestral bacterium. Thus while selective retention appears to have been the main determinant of the distribution of class I LysRSs, horizontal gene transfer has also played a significant but less widespread role.

Convergent Evolution of Class I and II LysRSs

Although the global and local architectures of LysRS-I and LysRS-II are fundamentally different, and while the two LysRSs approach the tRNA acceptor helix from opposite sides, their strategies to recognize tRNALys, as well as L-lysine, have considerable similarities. This strongly suggests functional convergence of the two unrelated enzymes and represents a typical example of convergent evolution. It also implies that tRNALys identity may have evolved prior to the enzyme specificity. The functional convergence of the two enzymes, with respect to the recognition of tRNALys and L-lysine, is illustrated by the mirror images of the LysRS-I and LysRS-II structures.20 Intriguingly, LysRS-I and LysRS-II can be docked simultaneously onto the tRNALys molecule without any steric clashes20 and it has recently been shown in vitro that both enzymes can bind tRNALys to form a stable ternary complex.16 Further characterization of this LysRS-I:tRNALys:LysRS-II ternary complex will provide insights into why two unrelated LysRSs exist, and may also impact our understanding of the development of the genetic code.5

Note Added in Proof

Recent studies have shown significant differences in noncognate amino acid recognition between class I and class II LysRS.35,36


Eriani G, Dirheimer G, Gangloff J. Cysteinyl-tRNA synthetase: determination of the last E. coli aminoacyl- tRNA synthetase primary structure. Nucleic Acids Res. 1991;19:265–269. [PMC free article: PMC333589] [PubMed: 2014166]
Arnez JG, Moras D. Structural and functional considerations of the aminoacylation reaction. Trends Biochem Sci. 1997;22:211–216. [PubMed: 9204708]
Cusack S. Aminoacyl-tRNA synthetases. Curr Opin Struct Biol. 1997;7:881–889. [PubMed: 9434910]
Ribas de Pouplana L, Schimmel P. Two classes of tRNA synthetases suggested by sterically compatible dockings on tRNA acceptor stem. Cell. 2001;104:191–193. [PubMed: 11269237]
Ribas de Pouplana L, Schimmel P. Aminoacyl-tRNA synthetases: potential markers of genetic code development. Trends Biochem Sci. 2001;26:591–596. [PubMed: 11590011]
Wilcox M, Nirenberg M. Transfer RNA as a cofactor coupling amino acid synthesis with that of protein. Proc Natl Acad Sci U S A. 1968;61:229–236. [PMC free article: PMC285927] [PubMed: 4972364]
Ibba M, Söll D. The renaissance of aminoacyl-tRNA synthesis. EMBO Rep. 2001;2:382–387. [PMC free article: PMC1083889] [PubMed: 11375928]
Tumbula DL, Becker HD, Chang WZ. et al. Domain-specific recruitment of amide amino acids for protein synthesis. Nature. 2000;407:106–110. [PubMed: 10993083]
Bult CJ, White O, Olsen GJ. et al. Complete genome sequence of the methanogenic archaeon, Methanococcus jannaschii. Science. 1996;273:1058–1073. [PubMed: 8688087]
Smith DR, Doucette-Stamm LA, Deloughery C. et al. Complete genome sequence of Methanobacterium thermoautotrophicum deltaH: functional analysis and comparative genomics. J Bacteriol. 1997;179:7135–7155. [PMC free article: PMC179657] [PubMed: 9371463]
Stathopoulos C, Li T, Longman R. et al. One polypeptide with two aminoacyl-tRNA synthetase activities. Science. 2000;287:479–482. [PubMed: 10642548]
Lipman RS, Sowers KR, Hou YM. Synthesis of cysteinyl-tRNACys by a genome that lacks the normal cysteine-tRNA synthetase. Biochemistry. 2000;39:7792–7798. [PubMed: 10869184]
Fabrega C, Farrow MA, Mukhopadhyay B. et al. An aminoacyl tRNA synthetase whose sequence fits into neither of the two known classes. Nature. 2001;411:110–114. [PubMed: 11333988]
Ibba M, Morgan S, Curnow AW. et al. A euryarchaeal lysyl-tRNA synthetase: resemblance to class I synthetases. Science. 1997;278:1119–1122. [PubMed: 9353192]
Ibba M, Bono JL, Rosa PA. et al. Archaeal-type lysyl-tRNA synthetase in the Lyme disease spirochete Borrelia burgdorferi. Proc Natl Acad Sci U S A. 1997;94:14383–14388. [PMC free article: PMC24988] [PubMed: 9405621]
Polycarpo C, Ambbrogelly A, Ruan B. et al. Activation of the pyrolysine suppressor tRNA requires formation of a ternary complex with class I and class II lysyl-tRNA synthetases. Mol Cell. 2003;12:287–294. [PubMed: 14536069]
Srinivasan G, James CM, Krzycki JA. Pyrrolysine encoded by UAG in Archaea: charging of a UAG-decodong specialized tRNA. Science. 2002;296:1459–1462. [PubMed: 12029131]
Hao B, Gong W, Ferguson TK. et al. A novel UAG encoded residue in the structure of a methanogen methyltransferase. Science. 2002;296:1462–1466. [PubMed: 12029132]
Ibba M, Losey HC, Kawarabayasi Y. et al. Substrate recognition by class I lysyl-tRNA synthetases: a molecular basis for gene displacement. Proc Natl Acad Sci U S A. 1999;96:418–423. [PMC free article: PMC15151] [PubMed: 9892648]
Terada T, Nureki O, Ishitani R. et al. Functional convergence of two lysyl-tRNA synthetases with unrelated topologies. Nat Struct Biol. 2002;9:257–262. [PubMed: 11887185]
Söll D, Becker HD, Plateau P. et al. Context-dependent anticodon recognition by class I lysyl-tRNA synthetases . Proc Natl Acad Sci U S A. 2000;97:14224–14228. [PMC free article: PMC18899] [PubMed: 11121028]
Sundaram M, Durant PC, Davis DR. Hypermodified nucleosides in the anticodon of tRNALys stabilize a canonical U-turn structure. Biochemistry. 2000;39:12575–12584. [PubMed: 11027137]
Stuart JW, Gdaniec Z, Guenther R. et al. Functional anticodon architecture of human tRNALys 3 includes disruption of intraloop hydrogen bonding by the naturally occurring amino acid modification, t6A. Biochemistry. 2000;39:13396–13404. [PubMed: 11063577]
Sylvers LA, Rogers KC, Shimizu M. et al. A 2-thiouridine derivative in tRNAGlu is a positive determinant for aminoacylation by Escherichia coli glutamyl-tRNA synthetase. Biochemistry. 1993;32:3836–3841. [PubMed: 8385989]
Ambrogelly A, Korencic D, Ibba M. Functional annotation of class I lysyl-tRNA synthetase phylogeny indicates a limited role for gene transfer. J Bacteriol. 2002;184:4594–4600. [PMC free article: PMC135231] [PubMed: 12142429]
Commans S, Lazard M, Delort F. et al. tRNA anticodon recognition and specification within subclass IIb aminoacyl-tRNA synthetases. J Mol Biol. 1998;278:801–813. [PubMed: 9614943]
Stello T, Hong M, Musier-Forsyth K. Efficient aminoacylation of tRNALys,3 by human lysyl-tRNA synthetase is dependent on covalent continuity between the acceptor stem and the anticodon domain. Nucleic Acids Res. 1999;27:4823–4829. [PMC free article: PMC148784] [PubMed: 10572184]
Nureki O, Vassylyev DG, Katayanagi K. et al. Architectures of class-defining and specific domains of glutamyl-tRNA synthetase. Science. 1995;267:1958–1965. [PubMed: 7701318]
Sekine S, Nureki O, Shimada A. et al. Structural basis for anticodon recognition by discriminating glutamyl- tRNA synthetase. Nat Struct Biol. 2001;8:203–206. [PubMed: 11224561]
Onesti S, Desogus G, Brevet A. et al. Structural studies of lysyl-tRNA synthetase: conformational changes induced by substrate binding. Biochemistry. 2000;39:12853–12861. [PubMed: 11041850]
Sekine S, Nureki O, Sakamoto K. et al. Major identity determinants in the “augmented D helix” of tRNAGlu from Escherichia coli. J Mol Biol. 1996;256:685–700. [PubMed: 8642591]
Ribas de Pouplana L, Turner RJ, Steer BA. et al. Genetic code origins: tRNAs older than their synthetases? Proc Natl Acad Sci U S A. 1998;95:11295–11300. [PMC free article: PMC21636] [PubMed: 9736730]
Wolf YI, Aravind L, Grishin NV. et al. Evolution of aminoacyl-tRNA synthetases—analysis of unique domain architectures and phylogenetic trees reveals a complex history of horizontal gene transfer events. Genome Res. 1999;9:689–710. [PubMed: 10447505]
Woese CR, Olsen GJ, Ibba M. et al. Aminoacyl-tRNA synthetases, the genetic code, and the evolutionary process. Microbiol Mol Biol Rev. 2000;64:202–236. [PMC free article: PMC98992] [PubMed: 10704480]
Jester B, Levengood J, Roy H. et al. Non-orthologous replacement of lysyl-tRNA synthetase prevents addition of lysine analogs to the genetic code. Proc Natl Acad Sci USA. 2003;100:14351–14356. [PMC free article: PMC283595] [PubMed: 14623972]
Levengood J, Ataide SF, Roy H. et al. Divergence in noncognate amino acid recognition between class I and class II lysyl-tRNA synthetases. J Biol Chem. 2004;279:17707–17714. [PubMed: 14747465]
Copyright © 2000-2013, Landes Bioscience.
Bookshelf ID: NBK6444


  • PubReader
  • Print View
  • Cite this Page

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...