Undetected antisense tRNAs in mitochondrial genomes?

Abstract

Background: The hypothesis that both mitochondrial (mt) complementary DNA strands of tRNA genes code for tRNAs (sense-antisense coding) is explored. This could explain why mt tRNA mutations are 6.5 times more frequently pathogenic than in other mt sequences. Antisense tRNA expression is plausible because tRNA punctuation signals mt sense RNA maturation: both sense and antisense tRNAs form secondary structures potentially signalling processing. Sense RNA maturation processes by default 11 antisense tRNAs neighbouring sense genes. If antisense tRNAs are expressed, processed antisense tRNAs should have adapted more for translational activity than unprocessed ones. Four tRNA properties are examined: antisense tRNA 5' and 3' end processing by sense RNA maturation and its accuracy, cloverleaf stability and misacylation potential.

Results: Processed antisense tRNAs align better with standard tRNA sequences with the same cognate than unprocessed antisense tRNAs, suggesting less misacylations. Misacylation increases with cloverleaf fragility and processing inaccuracy. Cloverleaf fragility, misacylation and processing accuracy of antisense tRNAs decrease with genome-wide usage of their predicted cognate amino acid.

Conclusions: These properties correlate as if they adaptively coevolved for translational activity by some antisense tRNAs, and to avoid such activity by other antisense tRNAs. Analyses also suggest previously unsuspected particularities of aminoacylation specificity in mt tRNAs: combinations of competition between tRNAs on tRNA synthetases with competition between tRNA synthetases on tRNAs determine specificities of tRNA amino acylations. The latter analyses show that alignment methods used to detect tRNA cognates yield relatively robust results, even when they apparently fail to detect the tRNA's cognate amino acid and indicate high misacylation potential.

Figure 1

Scheme of the human mitochondrial genome region templating for sense tRNAs Ser UGA and Asp GUC. Continuous lines indicate expressed sense genes, dashed lines hypothetically expressed antisense genes. tRNAs are indicated by their cognate amino acid, followed by the anticodon detected by tRNAscan-SE http://lowelab.ucsc.edu/tRNAscan-SE/ and the structural component of the Cove index of that tRNA. Hypothetical antisense tRNAs are in italics. Gene positions on the mitochondrial genome are indicated according to the standard in Genbank (NC_012920). The 5' flank of the antisense of tRNA Ser UGA is processed by sense RNA maturation by default because it is flanked by the mitochondrial protein coding gene COX1, its 3' flank because of its vicinity with sense tRNA Asp GUC. Normal sense RNA maturation processes only the 3' flank of the hypothetical antisense of sense tRNA Asp GUC. Extremities of tRNAs and their anticodons are as detected for these sequences by tRNAscan-SE.

Figure 2

Frequency distribution of misacylation tendency for sense and antisense tRNAs (white and black columns, respectively). The modal (most frequent) human tRNA sequence was used for each of the 22 mitochondrial tRNA species. TFAM aligns each sequence to reference tRNA sequences with known cognates. Alignment quality with each tRNA reference estimates the tendency of the focal tRNA sequence for acylation with the cognate of the tRNA reference (raw data in Table 2). The x axis indicates the number of tRNA references with cognates differing from the focal tRNA's cognate but aligning better with the focal tRNA sequence than the reference tRNA with the cognate matching the focal tRNA's anticodon. Hence numbers close to zero on the x axis indicate low tendency for misacylation. The y axis is the number of focal tRNAs observed for that x axis. The distribution expected according to a binomial distribution is also shown (see text for further explanations).

Figure 3

Misacylation potential as a function of antisense tRNA cloverleaf stability for the antisense of tRNA Tyr. The y axis is the number of amino acids with greater aminoacylation potential than serine, which is the predicted cognate for that antisense tRNA. The x axis is the residual of antisense COVE, from the linear regression between antisense COVE (dependent) with sense COVE (independent). Only taxa where sense-anticodon symmetry exists are considered, species are followed by NCBI accession numbers: Aotus lemurinus, FJ 85421; A. trivirgatus, AY 250707; Ateles belzebuth, FJ 785421; Callicebus donacophilus, FJ 785423; Cebus albifrons, NC 002763; Colobus guereza, NC 006901; Daubentonia madagascariensis, NC 010299; Eulemur fulvus, NC 012766; Eulemur mayottenis, NC 012769; Galago senegalensis, NC 012761; Gorilla gorilla, NC 001645; G. gorilla gorilla, NC 011120; Homo sapiens, NC 012920; Homo neanderthalensis, NC 011137; Hylobates lar, NC 002082; Lemur catta, NC 004025; Loris tardigradus, NC 012763; Macaca fascicularis, NC 012670; M. mulatta, NC 005943; M. sylvanus, NC 002764; M. thibetana, NC 011519; Nasalis larvatus, NC 008216; Nycticebus coucang, NC 002765; Otolemur crassicaudatus, NC 012762; Pan paniscus, NC 001644; P. troglodytes, NC 001643; Papio hamadryas, NC 001992; Pygathrix roxellana, NC 008218; Presbytis melalophos, NC 008217; Procolobus badius, NC 009219; Propithecus coquereli, NC 011053; Saguinus oedipus, FJ 785424; Semnopithecus entellus, NC 008215; Tarsius bancanus, NC 0021811; Theropithecus gelada, FJ 785426; Trachypithecus obscurus, NC 006900; Varecia varecia, NC 012773. Species explored but not included because of lack of anticodon symmetry in that specific tRNA are: Chlorocebus aethiops, NC 007009; C. pygerythrus, NC 009747; C. sabaeus, NC 008066; C. tantalus, NC 009748; Eulemur macaco, NC 012771; E. mongoz, NC 010300; Perodicticus potto, NC 012764; Pygathrix nemaeus, NC 008220; Pongo abelii, NC 002083; Pongo pygmaeus, NC 001646; Saimiri sciureus, NC 012775; Tarsius syrichta, NC 012774.

Figure 4

Misacylation potential of antisense of tRNA Trp as a function of processing inaccuracy of its 5' extremity by sense RNA maturation. The x axis is the number of nucleotides between the next sense gene (tRNA Ala) and the 5' extremity of the antisense of tRNA Trp. Datapoints are for the same species as in Figure 3, but species excluded for lack of anticodon symmetry between sense and antisense tRNAs differ.

Figure 5

Cloverleaf stability of antisense of tRNA Pro as a function of the accuracy of processing of its 3' extremity by sense RNA maturation. The x axis is the number of nucleotides between the next sense gene (tRNA Thr) and the 3' extremity of the antisense of tRNA Pro. Datapoints are for the same species as in Figure 3, but species excluded for lack of anticodon symmetry between sense and antisense tRNAs differ.

Figure 6

Column-versus row-analysis of TFAM's output. The y axis indicates the percentage of non-cognate amino acids with greater amino acylation potential than the cognate according to TFAM, assuming competition among tRNAs for tRNA synthetases (column-analysis of output in Table 2, see text). The x-axis is the percentage of non-cognate amino acids with greater aminoacylation potential than the cognate supposing competition among tRNA synthetases for tRNAs (row-analysis of Table 2). This graph shows that for many sense tRNAs for which classical interpretations of TFAM's output (row analysis of Table 2) yield a poor prediction of the cognate, assuming that tRNAs compete for tRNA synthetases explains apparent high misacylation rates.

Figure 7

Tendency for correct amino acylation in sense and antisense tRNAs as a function of duration of gestation in primates. The y axis plots numbers of tRNAs per mitochondrial genome with less than half non-cognate amino acids with aminoacylation scores (according to TFAM) higher than the tRNA's cognate predicted according to its anticodon. The x axis is the length of the gestation period (days), for primates (gestation data from http://genomics.senescence.info/species/[58]). Triangles: sense tRNAs; Circles, antisense tRNAs. The negative trend for antisense tRNAs suggests that antisense tRNAs are more adapted for translational activity in species with fast development than those with slower development. Data for sense tRNAs are presented as negative control.

Figure 8

Tendency for correct amino acylation in sense and antisense tRNAs as a function of duration of gestation in rodents. The y axis plots numbers of tRNAs per mitochondrial genome with less than half non-cognate amino acids with aminoacylation scores (according to TFAM) higher than the tRNA's cognate predicted according to its anticodon, as a function of the length of the gestation period (days), for rodents (gestation data from http://genomics.senescence.info/species/[58]). Triangles: sense tRNAs; Circles, antisense tRNAs. Results confirm those from Figure 7 for primates and suggest that the association with the duration of gestation is not circumstantial.