Aminoacylation properties of pathology-related human mitochondrial tRNA(Lys) variants

Abstract

In vitro transcription has proven to be a successful tool for preparation of functional RNAs, especially in the tRNA field, in which, despite the absence of post-transcriptional modifications, transcripts are correctly folded and functionally active. Human mitochondrial (mt) tRNA(Lys) deviates from this principle and folds into various inactive conformations, due to the absence of the post-transcriptional modification m(1)A9 which hinders base-pairing with U64 in the native tRNA. Unavailability of a functional transcript is a serious drawback for structure/function investigations as well as in deciphering the molecular mechanisms by which point mutations in the mt tRNA(Lys) gene cause severe human disorders. Here, we show that an engineered in vitro transcribed "pseudo-WT" tRNA(Lys) variant is efficiently recognized by lysyl-tRNA synthetase and can substitute for the WT tRNA as a valuable reference molecule. This has been exploited in a systematic analysis of the effects on aminoacylation of nine pathology-related mutations described so far. The sole mutation located in a loop of the tRNA secondary structure, A8344G, does not affect aminoacylation efficiency. Out of eight mutations located in helical domains converting canonical Watson-Crick pairs into G-U pairs or C.A mismatches, six have no effect on aminoacylation (A8296G, U8316C, G8342A, U8356C, U8362G, G8363A), and two lead to drastic decreases (5000- to 7000-fold) in lysylation efficiencies (G8313A and G8328A). This screening, allowing for analysis of the primary impact level of all mutations affecting one tRNA under comparable conditions, indicates distinct molecular origins for different disorders.

FIGURE 1.

Human mitochondrial wild-type and pseudo-WT tRNA^Lys transcripts. (A) Determined secondary structures of human mt wild-type tRNA^Lys transcript (Kwt; Helm et al. 1998). (B) Experimental secondary cloverleaf structures of native mt tRNA^Lys (displaying the six post-transcriptional modifications, abbreviated in the conventional way; Sprinzl et al. 1998) and of a chimeric tRNA^Lys methylated at position N1 of A9 as sole post-transcriptional modification (Helm et al. 1999b). (C) Pseudo-WT in vitro transcribed tRNA^Lys derivatives Ke and Ki. tRNA domains are labeled for Ke. Structural mutations introduced at positions 50 and 64 (indicated by gray circles) prevent alternate folding into extended hairpin (black crosses; Helm et al. 1998).

FIGURE 2.

From inactive to active in vitro transcribed human mt tRNA^Lys. Charging levels, determined with human LysRS at 37°C, are expressed as percentages of charged tRNA versus total amount of specific tRNA per experiment. Rz-Kwt, Rz-Ke, and Rz-Ki correspond to in vitro transcribed wild-type and structural variants of tRNA^Lys, respectively, purified to single nucleotide resolution on polyacryl-amide/urea gels. Purified native wild-type mt tRNA^Lys (by hybridization to a specific biotinylated oligonucleotide) form the positive control.

FIGURE 3.

Effect of MERRF (A8344G) mutation on aminoacylation. Charging levels of wild type or pseudo-WT (Rz-Ke and Rz-Ki) versus MERRF or pseudo-MERRF (Rz-Kem and Rz-Kim) tRNAs. Native wild-type and MERRF-mutated mt tRNA^Lys were extracted from mitochondria of R2–1A and R1–C3 cybrid cells (see Experimental Procedures) using biotinylated oligonucleotides complementary to the 3′ part of the tRNA. Rz- stands for self-cleaved in vitro transcribed tRNAs.

FIGURE 4.

Effect of MERRF mutation on the structure of in vitro transcribed tRNAs. (A) Autoradiographs of 15% denaturing polyacrylamide gels of chemical probing experiments. G indicates RNase T1 ladder; L, alkaline ladder; and C, control incubation without probe. (Top) Structural probing experiments with lead acetate (Pb(OAc)₂). 5′-Labeled tRNAs are incubated at 25°C in 20 μL of 40 mM Tris-HCl (pH 7.5), 10 mM MgCl₂, 40 mM NaCl, and 0.1 μg/μL unlabeled tRNA without (C) or with 0.15 mM Pb(OAc)₂ for 10 min (1), 20 min (2), or 30 min (3). (Bottom) Chemical probing with diethylpyrocarbonate (DEPC). N, SD, and D stand for native, semi-denaturating, and denaturating conditions, respectively. 5′-Labeled tRNAs are incubated for 40 min without (C) or with 5% (v/v) DEPC (N, SD, D) in 50 mM sodium cacodylate with 10 mM MgCl₂ and 300 mM KCl (C, N) or with 1 mM EDTA (SD, D). Incubation was performed for 4 min at 60°C for denaturating condition (D). (B) Chemical probing patterns on cloverleaf structures of pseudo-WT (Ke and Ki) and pseudo-MERRF (Kem and Kim) transcripts. Structural mutations (at position 50–64) and MERRF mutation (at position 55) are highlighted in bold. Arrowheads correspond to positions cleaved by probes (Pb(OAc)₂ and DEPC). Cleavages are categorized weak (white), modest (gray), or strong (black).

FIGURE 5.

Investigated mutations in human mitochondrial tRNA^Lys. Individual variants were prepared as derivatives of Rz-Ki, the pseudo-WT tRNA^Lys with U50–A64 pair instead of A50–U64 pair. The 10 reported pathological mutations (http://www.mitomap.org/) are indicated in black; the dashed arrow represents the one not investigated because it overlaps the structural mutations. Control molecules, indicated in gray, are either the polymorphic A8348G mutation (positive control) or the U8324A sequence variation (which alters a major lysine identity element, negative control). Numbering is either according to the mt genome location (bold characters, for mutations; Anderson et al. 1981) or according to the conventional tRNA annotation (small numbers; Sprinzl et al. 1998). Related pathologies are abbreviated as follows: DEAF, maternally inherited deafness or aminoglycoside-induced deafness; DMDF, diabetes mellitus and deafness; EM, encephalomyopathy; LS, Leigh syndrome; MELAS, mitochondrial encephalomyopathy, lactic acidosis, stroke-like episodes; MERRF, myoclonic epilepsy and ragged red fibers; MICM, maternal inherited cardiomyopathy; MNGIE, mitochondrial neurogastrointestinal encephalomyopathy; MS, myoclonic seizures; PEO, progressive external ophthalmoplegia; and SM, skeletal myopathy.

FIGURE 6.

Structural insight on pathology-related mutations. (A) Summary of the effects of mutations on tRNA^Lys aminoacylation efficiencies. Values are from Table 1 ▶ and the color code is as follows: structural mutations introduced at positions 50 and 64, which prevent alternate folding into extended hairpin (Helm et al. 1998), are indicated in green. Polymorphic mutation A8348G is in yellow. Neutral, mild, and highly deleterious mutations are, respectively, in blue, pink, and magenta. (B) Sequence conservation of mammalian mt tRNA^Lys genes (Helm et al. 2000) and location of pathology-related mutations studied in the present work (arrowheads). The color code for arrowheads is as in A. (C–E) Location of mutations in three-dimensional representations of the tRNA. The color code is the same as in parts A and B. Nucleotides complementary to mutated positions and thus involved in base-pairing are in white. The model tRNA is a ribbon representation of the crystal structure of yeast tRNA^Asp (in its complexed form with AspRS; Ruff et al. 1991; Cavarelli et al. 1993). (E) Profile representation emphasizing the location of harmful mutations within a same face of the tRNA. The CCA end is pointed toward the reader.