A complete landscape of post-transcriptional modifications in mammalian mitochondrial tRNAs

Abstract

In mammalian mitochondria, 22 species of tRNAs encoded in mitochondrial DNA play crucial roles in the translation of 13 essential subunits of the respiratory chain complexes involved in oxidative phosphorylation. Following transcription, mitochondrial tRNAs are modified by nuclear-encoded tRNA-modifying enzymes. These modifications are required for the proper functioning of mitochondrial tRNAs (mt tRNAs), and the absence of these modifications can cause pathological consequences. To date, however, the information available about these modifications has been incomplete. To address this issue, we isolated all 22 species of mt tRNAs from bovine liver and comprehensively determined the post-transcriptional modifications in each tRNA by mass spectrometry. Here, we describe the primary structures with post-transcriptional modifications of seven species of mt tRNAs which were previously uncharacterized, and provide revised information regarding base modifications in five other mt tRNAs. In the complete set of bovine mt tRNAs, we found 15 species of modified nucleosides at 118 positions (7.48% of total bases). This result provides insight into the molecular mechanisms underlying the decoding system in mammalian mitochondria and enables prediction of candidate tRNA-modifying enzymes responsible for each modification of mt tRNAs.

Figure 1.

Mass spectrometric analysis of bovine mt tRNA^Ala for assignment of post-transcriptional modifications. (A) Secondary structure of bovine mitochondrial tRNA^Ala with post-transcriptional modifications determined in this study. The position numbers of the modifications are displayed according to the nucleotide numbering system from the tRNA compilation (40). Symbols for modified nucleosides are as follows: m¹A, 1-methyladenosine; m²G, N²-methylguanosine and Ψ, pseudouridine. Watson–Crick base pairs are indicated by solid lines, whereas G–U pairs are indicated by asterisks. (B) Nucleoside analysis of bovine mitochondrial tRNA^Ala. Left, top panel: UV chromatogram at 254 nm of the four major nucleosides (C, U, G and A). Left, lower panels: extracted-ion chromatograms (XIC) for the protonated ion of m¹A nucleoside (m/z 282, black line) with its base ion (m/z 150, gray line) (second panel), m²G nucleoside (m/z 298, black line) with its base ion (m/z 166, gray line) (third panel) and Ψ nucleoside (m/z 245, black line) (bottom panel). The XIC for the base ion (20% upper offset) is overlaid on the XIC for the nucleoside ion. Right: mass spectra of m¹A and m²G. Cleavage positions for the base-related ions are indicated on the chemical structures. (C) RNA fragment analysis of RNase T₁ digests of bovine mitochondrial tRNA^Ala. Assigned fragments are indicated on the base peak chromatogram (BPC) in the first panel. ‘p’ stands for the terminal phosphate group. The XIC for the doubly charged negative ion of a modification-containing fragment (AUUUm¹Am²Gp, m/z 982.6) is indicated in the second panel. Because m²G at position 10 is a partial modification, both AUUUm¹Am²Gp and AUUUm¹AGp were detected. (D) RNA fragment analysis of RNase A digests of bovine mitochondrial tRNA^Ala. Assigned fragments are indicated on the BPC. The XIC for the doubly charged negative ion of the 5′-terminal fragment (pGAGGAUp, m/z 1047.6) is indicated in the second panel. (E) A CID spectrum of a cyanoethylated RNA fragment to determine the location of a Ψ site. The doubly charged negative ion of the RNA fragment (m/z 1617.7) shown in the inset was used as the precursor ion for CID. The product ions were assigned according to McLuckey et al. (41). The asterisks in the spectrum denote product ions containing ce¹Ψ. (F) Whole mass analysis of intact bovine mt tRNA^Ala. A series of multiply charged negative ions is shown in the mass spectrum. The charge values are indicated in parentheses. The observed mass obtained by deconvoluting the mass spectra is shown in the inset.

Figure 2.

Post-transcriptional modifications in six bovine mt tRNAs. Symbols for modified nucleosides are as follows: m¹G, 1-methylguanosine; i⁶A, N⁶-isopentenyladenosine; Q, queuosine; m⁵C, 5-methylcytidine; ms²i⁶A, 2-methylthio-N⁶-isopentenyladenosine. The ‘G/A’ in tRNA^Asp and tRNA^Tyr indicates that both G and A were detected at this position, probably as a result of heteroplasmy in the bovine liver we used in this study.

Figure 3.

Revised information regarding post-transcriptional modifications of five bovine mt tRNAs. The updated modified bases are represented in bold type. Symbols for modified nucleoside are as follows: m²₂G, N², N²-dimethylguanosine; D, dihydrouridine; τm⁵U, 5-taurinomethyluridine; τm⁵s²U, 5-taurinomethyl-2-thiouridine; t⁶A, N⁶-threonylcarbamoyladenosine; m³C, 3-methylcytidine.

Figure 4.

Summary of post-transcriptional modifications in bovine mt tRNAs. Species and numbers of post-transcriptional modifications identified in 22 bovine mt tRNAs are mapped on the schematic secondary structure of tRNA. The modified positions are depicted by gray circles with a symbol indicating each modification. At each position, the number of tRNAs that bear the modification are shown in parenthesis. Positions 27a and 43a, indicated by light gray circles, are unique to mt tRNA^Ser(UCN). G_-1 is specific to mt tRNA^His.