Some cool stuff about mitochondrial translation… Lots of links to good papers at the end (in what someone pointed out is a Bibel-iography at the end) for those wanting more. But here are some of my notes while reading them

human mt-mRNAs are weird… 

  • most* don’t have 5’ & 3’ UTRs – instead they use “leaderless translation” where the start codon is just right at the 5’ end
  • they don’t have 5’ 7-methylguanosine caps like cytoplasmic eukaryotic mRNAs do
  • they don’s have Shine-Delgarno sequences (SDSs) like bacterial mRNAs often do
  • some of them are transcribed as polycistronic transcripts that get processed into individual mt-mRNAs by cutting out the tRNAs separating them
    • this prevents the need for splicing
  • there are only 11 of them (encoding 13 proteins, but 2 of the mRNAs are bicistronic (they have 2 proteins encoded in them back-to-back)

*CO1 has 3 nucleotides (nt)(RNA letters) before its start; ND1 has 2 nt before its start; and ATP8 has 1 nt before its start – but these are still WAY shorter than cytoplasmic 5’ UTRs

*ATP8/6 & MTND4L/4 (those bicistronic ones) *do* have a leader sequence (at least the second one does) because they have another protein’s instructions in front of them!

similarly to our cytosolic mRNAs, they do get 3’ polyadenylated (well, all but one of them)- this is done by mitochondrial poly(A) polymerase (PAPD)

  • a cool thing about this is that, for 7 of them, the extra A’s needed to complete the stop codon
  • this polyadenylation is promoted by this protein complex called LRPPRC/SLIRP, which relaxes the mRNAs’ secondary structure so PAPD can do its job
  • if you’re wondering about that “all but one” thing, apparently ND6 isn’t poly-A-ed

it’s not just the mt-mRNA that’s weird…

the mt-tRNAs are weird too!

  • first off, there are fewer of them – there are “only” 22 (and they’re made from mitochondrial DNA), compared to the 40-60-ish cytoplasmic ones
    • they get post-transcriptionally modified in the anticodon region (which things like tuaurinomethyl added to uridine (e.g. τm5U) & sulfur added to uridine (e.g. τm5s2U)  let them recognize more synonymous codons (3-letter RNA words that spell the same amino acid) via wobble base pairing (i.e. you don’t need strict canonical A-U, C-G pairing)
      • note:they also get modified at other positions for stability, efficiency, etc.
  • their structures are more diverse, including some with weird truncations

they use an altered genetic code, “reassigning” codons used in the “universal” genetic code:

  • “normally” AUA spells isoleucine (Ile), but in the mitochondria it spells methionine (Met)
  • “normally” UGA serves as a stop codon, but in the mitochondria it spells tryptophan (Trp)
  • “normally” AGA & AGG spell arginine (Arg), but in the mitochondria they serve as stop codons

mitoribosomes themselves are weird

  • as I talked about in a previous post, they have more protein components and less rRNA
  • they’re smaller than our cytosolic ribosomes & bacterial ribosomes
  • they associate with the inner mitochondrial membrane so that (with some help from other proteins and lipids) they can insert the nascent peptide (the protein in the process of getting made) into the membrane so that the proteins (which are all membrane proteins) can fold properly in their native habitat
  • there’s a crucial part of bacterial ribosomes called 5S rRNA. Mitoribosomes have ditched it evolutionarily and replaced it with a tRNA!!!! (either mitochondrial valine or phenylalanine tRNA embedded in the ribosomal structure)
  • since mitochondria are derived from an ancient cell swallowing a prokaryotic cell, mitochondrial ribosomes are more “prokaryote-like” than our cytosolic ones are. And this can have important medical consequences because some bacterial ribosome-targeting antibiotics can also bind to and mess up the workings of mitochondrial ribosomes. This is especially an issue if people have sensitizing mutations in their mitochondrial rRNA (ribosomal RNA) that makes it more “bacteria-like” and susceptible to antibiotic-induced hearing loss.  
  • interestingly, mt-rRNA is less modified than other ribosomal rRNAs – for example, there are only 10 modified bases (see the Rebelo reference at the end) – and the modifications don’t require small nucleolar ribonucleoproteins (snoRNPs) like the cytoplasmic ones do
  • more in this past post: blog:; YouTube:  

mitochondrial translation can be tricky to study

  • they’re “locked up” in membrane-bound structures, so it’s hard to get exogenous nucleic acids to them to try to get them to make different things in vivo or in organello (in isolated mitochondria)
  • they’re a lot less abundant than cytosolic or bacterial ones
  • they’re membrane associated (as are the proteins they make)
  • they use those weird tRNAs
  • more on methods used to study it in the Apostolopoulos & Iwasaki review linked to below



Basil J. Greber and Nenad Ban (2016) Structure and Function of the Mitochondrial Ribosome .  Annual Review of Biochemistry   

Rackham, O., Filipovska, A. Organization and expression of the mammalian mitochondrial genome. Nat Rev Genet (2022). 

Rebelo-Guiomar, P., Powell, C. A., Van Haute, L., & Minczuk, M. (2019). The mammalian mitochondrial epitranscriptome. Biochimica et biophysica acta. Gene regulatory mechanisms1862(3), 429–446. 

Antonios Apostolopoulos, Shintaro Iwasaki, Into the matrix: current methods for mitochondrial translation studies, The Journal of Biochemistry, Volume 171, Issue 4, April 2022, Pages 379–387,

more on modifications:

Suzuki, T., Yashiro, Y., Kikuchi, I. et al. Complete chemical structures of human mitochondrial tRNAs. Nat Commun 11, 4269 (2020). 

Rebelo-Guiomar, P., Pellegrino, S., Dent, K.C. et al. A late-stage assembly checkpoint of the human mitochondrial ribosome large subunit. Nat Commun 13, 929 (2022). 


Aibara, S., Singh, V., Modelska, A., & Amunts, A. (2020). Structural basis of mitochondrial translation. eLife9, e58362.

Itoh, Y., Andréll, J., Choi, A., Richter, U., Maiti, P., Best, R. B., Barrientos, A., Battersby, B. J., & Amunts, A. (2021). Mechanism of membrane-tethered mitochondrial protein synthesis. Science (New York, N.Y.)371(6531), 846–849. 

Yuzuru Itoh, Anas Khawaja, Vivek Singh, Andreas Naschberger, Joanna Rorbach, Alexey Amunts. Structural basis of streptomycin off-target binding to human mitoribosome. bioRxiv 2022.02.02.478878; doi:

random cool stuff about ribosomes:  

note: thanks to David Cruz for pointing out the Bibel-iography thing

Leave a Reply

Your email address will not be published.