Modifications on mRNA offer the potential of regulating mRNA fate post-transcriptionally. Recent studies suggested the widespread presence of N-methyladenosine (mA), which disrupts Watson-Crick base pairing, at internal sites of mRNAs. These studies lacked the resolution of identifying individual modified bases, and did not identify specific sequence motifs undergoing the modification or an enzymatic machinery catalysing them, rendering it challenging to validate and functionally characterize putative sites. Here we develop an approach that allows the transcriptome-wide mapping of mA at single-nucleotide resolution. Within the cytosol, mA is present in a low number of mRNAs, typically at low stoichiometries, and almost invariably in tRNA T-loop-like structures, where it is introduced by the TRMT6/TRMT61A complex. We identify a single mA site in the mitochondrial ND5 mRNA, catalysed by TRMT10C, with methylation levels that are highly tissue specific and tightly developmentally controlled. mA leads to translational repression, probably through a mechanism involving ribosomal scanning or translation. Our findings suggest that mA on mRNA, probably because of its disruptive impact on base pairing, leads to translational repression, and is generally avoided by cells, while revealing one case in mitochondria where tight spatiotemporal control over mA levels was adopted as a potential means of post-transcriptional regulation.

Adenosine (A) to inosine (I) RNA editing is widespread in eukaryotes. In prokaryotes, however, A-to-I RNA editing was only reported to occur in tRNAs but not in protein-coding genes. By comparing DNA and RNA sequences of , we show for the first time that A-to-I editing occurs also in prokaryotic mRNAs and has the potential to affect the translated proteins and cell physiology. We found 15 novel A-to-I editing events, of which 12 occurred within known protein-coding genes where they always recode a tyrosine (TAC) into a cysteine (TGC) codon. Furthermore, we identified the tRNA-specific adenosine deaminase (tadA) as the editing enzyme of all these editing sites, thus making it the first identified RNA editing enzyme that modifies both tRNAs and mRNAs. Interestingly, several of the editing targets are self-killing toxins that belong to evolutionarily conserved toxin-antitoxin pairs. We focused on hokB, a toxin that confers antibiotic tolerance by growth inhibition, as it demonstrated the highest level of such mRNA editing. We identified a correlated mutation pattern between the edited and a DNA hard-coded Cys residue positions in the toxin and demonstrated that RNA editing occurs in in two additional bacterial species. Thus, not only the toxin is evolutionarily conserved but also the editing itself within the toxin is. Finally, we found that RNA editing in increases as a function of cell density and enhances its toxicity. Our work thus demonstrates the occurrence, regulation, and functional consequences of RNA editing in bacteria.

The mitochondrial ribosome, which translates all mitochondrial DNA (mtDNA)-encoded proteins, should be tightly regulated pre- and post-transcriptionally. Recently, we found RNA-DNA differences (RDDs) at human mitochondrial 16S (large) rRNA position 947 that were indicative of post-transcriptional modification. Here, we show that these 16S rRNA RDDs result from a 1-methyladenosine (m1A) modification introduced by TRMT61B, thus being the first vertebrate methyltransferase that modifies both tRNA and rRNAs. m1A947 is conserved in humans and all vertebrates having adenine at the corresponding mtDNA position (90% of vertebrates). However, this mtDNA base is a thymine in 10% of the vertebrates and a guanine in the 23S rRNA of 95% of bacteria, suggesting alternative evolutionary solutions. m1A, uridine, or guanine may stabilize the local structure of mitochondrial and bacterial ribosomes. Experimental assessment of genome-edited Escherichia coli showed that unmodified adenine caused impaired protein synthesis and growth. Our findings revealed a conserved mechanism of rRNA modification that has been selected instead of DNA mutations to enable proper mitochondrial ribosome function.

No abstract

RNA transcripts are generally identical to the underlying DNA sequences. Nevertheless, RNA–DNA differences (RDDs) were found in the nuclear human genome and in plants and animals but not in human mitochondria. Here, by deep sequencing of human mitochondrial DNA (mtDNA) and RNA, we identified three RDD sites at mtDNA positions 295 (C-to-U), 13710 (A-to-U, A-to-G), and 2617 (A-to-U, A-to-G). Position 2617, within the 16S rRNA, harbored the most prevalent RDDs (>30% A-to-U and ∼15% A-to-G of the reads in all tested samples). The 2617 RDDs appeared already at the precursor polycistrone mitochondrial transcript. By using traditional Sanger sequencing, we identified the A-to-U RDD in six different cell lines and representative primates (Gorilla gorilla, Pongo pigmaeus, and Macaca mulatta), suggesting conservation of the mechanism generating such RDD. Phylogenetic analysis of more than 1700 vertebrate mtDNA sequences supported a thymine as the primate ancestral allele at position 2617, suggesting that the 2617 RDD recapitulates the ancestral 16S rRNA. Modeling U or G (the RDDs) at position 2617 stabilized the large ribosomal subunit structure in contrast to destabilization by an A (the pre-RDDs). Hence, these mitochondrial RDDs are likely functional.

No abstract

Compatibility between the nuclear (nDNA) and mitochondrial (mtDNA) genomes is important for organismal health. However, its significance for major evolutionary processes such as speciation is unclear, especially in vertebrates. We previously identified a sharp mtDNA-specific sequence divergence between morphologically indistinguishable chameleon populations (Chamaeleo chamaeleon recticrista) across an ancient Israeli marine barrier (Jezreel Valley). Because mtDNA introgression and gender-based dispersal were ruled out, we hypothesized that mtDNA spatial division was maintained by mito-nuclear functional compensation. Here, we studied RNA-seq generated from each of ten chameleons representing the north and south populations and identified candidate nonsynonymous substitutions (NSSs) matching the mtDNA spatial distribution. The most prominent NSS occurred in 14 nDNA-encoded mitochondrial proteins. Increased chameleon sample size (N = 70) confirmed the geographic differentiation in POLRMT, NDUFA5, ACO1, LYRM4, MARS2, and ACAD9. Structural and functionality evaluation of these NSSs revealed high functionality. Mathematical modeling suggested that this mito-nuclear spatial divergence is consistent with hybrid breakdown. We conclude that our presented evidence and mathematical model underline mito-nuclear interactions as a likely role player in incipient speciation in vertebrates.

The Levant is an important migration bridge, harboring border-zones between Afrotropical and palearctic species. Accordingly, Chameleo chameleon, a common species throughout the Mediterranean basin, is morphologically divided in the southern Levant (Israel) into two subspecies, Chamaeleo chamaeleon recticrista (CCR) and C. c. musae (CCM). CCR mostly inhabits the Mediterranean climate (northern Israel), while CCM inhabits the sands of the north-western Negev Desert (southern Israel). AFLP analysis of 94 geographically well dispersed specimens indicated moderate genetic differentiation (PhiPT = 0.097), consistent with the classical division into the two subspecies, CCR and CCM. In contrast, sequence analysis of a 637 bp coding mitochondrial DNA (mtDNA) fragment revealed two distinct phylogenetic clusters which were not consistent with the morphological division: one mtDNA cluster consisted of CCR specimens collected in regions northern of the Jezreel Valley and another mtDNA cluster harboring specimens pertaining to both the CCR and CCM subspecies but collected southern of the Jezreel Valley. AMOVA indicated clear mtDNA differentiation between specimens collected northern and southern to the Jezreel Valley (PhiPT = 0.79), which was further supported by a very low coalescent-based estimate of effective migration rates. Whole chameleon mtDNA sequencing (∼17,400 bp) generated from 11 well dispersed geographic locations revealed 325 mutations sharply differentiating the two mtDNA clusters, suggesting a long allopatric history further supported by BEAST. This separation correlated temporally with the existence of an at least 1 million year old marine barrier at the Jezreel Valley exactly where the mtDNA clusters meet. We discuss possible involvement of gender-dependent life history differences in maintaining such mtDNA genetic differentiation and suggest that it reflects (ancient) local adaptation to mitochondrial-related traits.