Biosynthesis of wyosine derivatives in tRNA: an ancient and highly diverse pathway in Archaea

Abstract

Wyosine (imG) and its derivatives such as wybutosine (yW) are found at position 37 of phenylalanine-specific transfer RNA (tRNA(Phe)), 3' adjacent to the anticodon in Eucarya and Archaea. In Saccharomyces cerevisiae, formation of yW requires five enzymes acting in a strictly sequential order: Trm5, Tyw1, Tyw2, Tyw3, and Tyw4. Archaea contain wyosine derivatives, but their diversity is greater than in eukaryotes and the corresponding biosynthesis pathways still unknown. To identify these pathways, we analyzed the phylogenetic distribution of homologues of the yeast wybutosine biosynthesis proteins in 62 archaeal genomes and proposed a scenario for the origin and evolution of wyosine derivatives biosynthesis in Archaea that was partly experimentally validated. The key observations were 1) that four of the five wybutosine biosynthetic enzymes are ancient and may have been present in the last common ancestor of Archaea and Eucarya, 2) that the variations in the distribution pattern of biosynthesis enzymes reflect the diversity of the wyosine derivatives found in different Archaea. We also identified 7-aminocarboxypropyl-demethylwyosine (yW-86) and its N4-methyl derivative (yW-72) as final products in tRNAs of several Archaea when these were previously thought to be only intermediates of the eukaryotic pathway. We confirmed that isowyosine (imG2) and 7-methylwyosine (mimG) are two archaeal-specific guanosine-37 derivatives found in tRNA of both Euryarchaeota and Crenarchaeota. Finally, we proposed that the duplication of the trm5 gene in some Archaea led to a change in function from N1 methylation of guanosine to C7 methylation of 4-demethylwyosine (imG-14).

FIG. 1.

Guanosine derivatives identified in naturally occurring tRNA^Phe of various origins. The derivatives are listed in order of complexity. The symbols for wyosine derivatives are those used in the original papers of McCloskey et al. for Archaea (Zhou, Sitaramaiah, Noon, et al. 2004) or more recently by Noma et al. in Saccharomyces cerevisiae (Noma et al. 2006). The numbering of the atoms is different for guanosine (IUPAC convention) and for wyosine derivatives that follow the convention defined by Blobstein et al. (1973) that was systematically used in recent papers on wyosine nucleosides. In panels G and H, the acp group originating from S-AdoMet is boxed

FIG. 2.

Evolution of the enzymes involved in wyosine biosynthetic pathway in Archaea. The phylogeny shown on the left is a schematic drawing of the relationships between the main archaeal lineages (most of the representative 62 organisms of which fully sequenced genomes were available last July 2009) based on previous published works (Brochier-Armanet et al. 2008). LACA corresponds to the last archaea common ancestor. Depending on the evolutionary scenario considered, LACA may be also the ancestor of Eucarya (see text for more details). The brown, yellow, and green filled circles represent Trm5 subgroups. Blue squares, orange circles, and pink triangles indicate the presence of Taw1, Taw2, or Taw3 homologues, respectively. Evolutionary events that occurred during the evolution of these proteins are indicated by letters: “H” indicates a putative HGT event and “+2” indicates a duplication event. Gene losses are symbolized by empty symbols.

FIG. 3.

Unrooted phylogenies of archaeal enzymes involved in wyosine derivative synthesis. (A) Trm5 (168 unambiguously aligned positions), (B) Taw1 (232 unambiguously aligned positions), (C) Taw2 (177 unambiguously aligned positions), and (D) Taw3 (140 unambiguously aligned positions). The topologies of these trees were computed by a Bayesian approach. Numbers ≤1 close to nodes represent the posterior probabilities, whereas those ≥1 correspond to ML BVs computed by Treefinder (for clarity only BVs greater than 50% are shown). The scale bar represents the average number of substitutions per site. Squares correspond to the five archaeal phyla: black, Crenarchaeota; black outlined in gray, Korarchaeota; white outlined in black, Thaurmarchaeota; white outlined in gray, Nanoarchaeota; and gray, Euryarchaeota. Group names ending in –ales represent orders.

FIG. 4.

Predicted biosynthetic pathways of archaeal wyosine derivatives. Acronym names for wyosine derivatives (boxed) and for enzymes (not boxed) are defined in figure 2 and also in text. Wyosine derivatives that are unique to the archaeal domain are in gray background. Symbols (colored circles, squares, and triangles) are defined in figure 2. Green squares with a FMN inside correspond to the flavodoxin domain that is needed to allow Tyw1 to catalyze the formation of imG-14. In Archaea, this domain apparently must act in trans and is yet to be identified. Symbol ‘?’ means that the enzyme has not been characterized. A black arrows means that the enzyme certainly catalyzes the reaction (when dashed the quantities of the product detected were small). A dashed red arrow corresponds to a predicted but not experimentally validated reaction.

FIG. 5.

Analysis of yW content in bulk tRNA from selected archaeal species. Chromatograms showing the separation of the yW derivatives from archaeal tRNA (Haloferax volcanii, Pyrococcus furiosus, Methanococcus maripaludis, Methanosarcina acetivorans, Pyroc. abyssi, Halobacterium salinarium, Sulfolobus shibatae, and Pyrobaculum calidifontis) using LC-MS/MS. The modifications detected and their retention times are indicated in the chromatograms. Next to each ultraviolet trace, the extracted ion chromatograms (XIC) of each yW derivative detected were also indicated. The peak area (PA) indicated next to the peaks was integrated from the XIC of each modification (m¹G, imG-14, imG, imG2, mimG, yW-86, and yW-72) and normalized using the m² ₂G modification (MH⁺ 312 m/z).

FIG. 6.

Domain organization of COG2520 proteins. A summary of the structure-based alignments presented in supplementary data 7, Supplementary Material online, is shown here. The dashed gray D1 domains signify that some members of subfamily have the domain and some not. Full black D1 domains mean all members of the family have one. Numbering in the motifs is based on the Methanocaldococcus jannaschii Trm5b sequence (MJ0883). The one letter conventional symbols for amino acids are used. h means hydrophobic residue and x means any residue, residues that are in more than 70% of the subfamily are underlined and residues conserved in all the subfamilies are boxed.