The role of tRNA identity elements in aminoacyl-tRNA editing

Abstract

The rules of the genetic code are implemented by the unique features that define the amino acid identity of each transfer RNA (tRNA). These features, known as "identity elements," mark tRNAs for recognition by aminoacyl-tRNA synthetases (ARSs), the enzymes responsible for ligating amino acids to tRNAs. While tRNA identity elements enable stringent substrate selectivity of ARSs, these enzymes are prone to errors during amino acid selection, leading to the synthesis of incorrect aminoacyl-tRNAs that jeopardize the fidelity of protein synthesis. Many error-prone ARSs have evolved specialized domains that hydrolyze incorrectly synthesized aminoacyl-tRNAs. These domains, known as editing domains, also exist as free-standing enzymes and, together with ARSs, safeguard protein synthesis fidelity. Here, we discuss how the same identity elements that define tRNA aminoacylation play an integral role in aminoacyl-tRNA editing, synergistically ensuring the correct translation of genetic information into proteins. Moreover, we review the distinct strategies of tRNA selection used by editing enzymes and ARSs to avoid undesired hydrolysis of correctly aminoacylated tRNAs.

Keywords: aminoacyl-tRNA synthetases; editing; mistranslation; protein synthesis; tRNA; translational fidelity.

FIGURE 1

(A) Steps in tRNA aminoacylation and editing. tRNAs are aminoacylated by ARSs producing aa-tRNAs. If the ARS uses a non-cognate amino acid (ncaa), the resulting ncaa-tRNA can be hydrolyzed by the editing enzymes. In the absence of editing checkpoints, the ncaa is incorporated into proteins in response to the wrong codon, causing mistranslation. (B) Representative secondary structures of tRNAs. As discussed in the main text, the numbered bases indicate the various positions important for editing. (C) Summary of the trans- and cis-editing domains with characterized functions and their known tRNA recognition elements. ^aArchaeal origin; ^bindicates weak or no tRNA specificity; ^cthe specificity of N73 depends on the DTD’s origin; ^din the context of tRNA^Ala; ^eBacterial origin; “ND” indicates not determined. B and E for ProXp-Ala indicate bacterial and eukaryotic, respectively.

FIGURE 2

Representative structures of ARSs’ editing domains (A) and free-standing editing enzymes (B). The CP1 domains of LeuRS (PDB 3ZJU), ValRS (PDB 1IVS), and IleRS (PDB 1FFY) are colored in light blue, teal, and navy blue, respectively. The editing domains of ThrRS (PDB 1NYQ), AlaRS (PDB 3WQY), ProRS (PDB 2J3L), and PheRS (PDB 3PCO) are shown in green, pink, orange, and purple, respectively. The other domains (e.g., aminoacylation and anticodon binding domains) are in black. For ThrRS-ed (an AlphaFold model of S. solfataricus), the hydrolytic active domain is shown in green, while the anticodon binding domain is in black. The structure of E. coli DTD (PDB 1JKE) and the AlphaFold model of human ATD are shown. The INS superfamily is represented by ProRS and five single-domain families: YbaK (PDB 1DBU), ProXp-ST1 (an AlphaFold model of E. coli), ProXp-ST2 (an AlphaFold model of Bordetella parapertussis), ProXp-ala (PDB 5VXB), ProXp-ala-CTD (an AlphaFold model of Arabidopsis thaliana), ProXp-ala-ProRS (an AlphaFold model of Plasmodium falciparum) and ProXp-x (PDB 2CX5). ProXp-7, ProXp-8, and ProXp-9 were omitted from the INS superfamily because their activities are unknown. The three known isoforms of AlaXp are represented by the structures of AlaXp-S (PDB 1WXO), AlaXp-M (PDB 2E1B), and AlaXp-L (an AlphaFold model of Pyrococcus horikoshii). For simplicity, all structures are displayed in monomeric form.