As detailed in Figure 1, the human mitochondrial genome contains just 37 genes. Whereas nuclear genes often have their own dedicated promoters, transcription of mtDNA results in large multigenic transcripts, just as in transcription of bacterial DNA. The large multigenic transcripts produced by transcribing the H- and L-strands are subsequently cleaved to generate a mix of coding and noncoding RNAs, as described below.
• Thirteen mRNAs. They are translated on mitoribosomes to make 13 protein com ponents of mitochondrial respiratory complexes, the membrane-bound enzymes of oxidative phosphorylation used to make ATP.
• Two rRNAs and 22 tRNAs. (Note that the mitochondrial tRNAVal has a dual role: as well as carrying valine to be incorporated into proteins, some of the tRNAVal molecules act as structural and functional components of the mitoribosome, serving a role analogous to that of 5S rRNA in cytoplasmic ribosomes.)
Fig1. Organization of the human mitochondrial genome. The circular 16,569 bp genome has a base composition of 44% GC, but has a heavy (H) strand rich in guanines, and a light (L) strand rich in cytosines. Twenty-four RNA genes make 12S and 16S rRNAs and 22 tRNAs (tRNA genes are shown as thin red bars with a letter indicating the amino acid; note there are two tRNALeu genes, L1 and L2 , and two tRNASer genes, S1 and S2 ). Thirteen protein-coding genes make components of the oxidative phosphorylation system: seven NADH dehydrogenase subunits (ND1–ND6 and ND4L), two ATP synthase subunits (ATP6 and ATP8), three cytochrome c oxidase subunits (CO1–CO3), and cytochrome b (CYB). Two promoters, green boxes labeled PH (in two segments) and PL , transcribe respectively the H- and L-strands in opposite directions, generating large multigenic transcripts from each strand that are subsequently cleaved. OH and OL (purple boxes) signify replication origins (dashed arrows mark the direction of DNA synthesis). A roughly 500 bp control region (CR), commencing at position 16024, has multiple regulatory sequences. It has an internal displacement loop (D-loop), a triple-stranded structure formed by repeat synthesis of a short piece of heavy-strand DNA, 7S DNA, that base pairs with the L-strand, causing local looping of the H-strand. Gene symbols for protein coding genes are abbreviated by omitting the prefix MT-. For further information, see the MITOMAP database at http:// www. mitomap.org.
Like bacterial genomes, the mitochondrial genome is extremely compact. In addition to lacking introns, the mitochondrial genes are very tightly packed. In most cases the sequences of neighboring genes are contiguous, or separated by just one or two noncoding bases, and two of the protein-coding genes, MT-ATP6 and MT-ATP8, have overlapping coding sequences with different reading frames. Space is con served to such an extent that some of the protein-coding genes even lack termination codons; to overcome this deficiency, UAA codons have to be introduced at the post- transcriptional level.
The limited autonomy of the mitochondrial genome
The mitochondrial genome has retained a measure of independence when it comes to protein synthesis because all the ribosomal RNAs and tRNAs needed for protein synthesis on mitoribosomes are made by transcribing mtDNA. However, it is a different matter for mitochondrial proteins. Only 13 out of the 80 proteins required for oxidative phosphorylation are specified by the mitochondrial genome (Table 1). None of the core nucleoid proteins—including critically important proteins needed to replicate, transcribe, and repair mtDNA—are specified by the mitochondrial genome. And although all the RNAs needed for translating mitochondrial mRNAs are made by mtDNA, all the proteins of the protein-synthesis machinery (mitochondrial ribosomal proteins and aminoacyl tRNA synthetases) are made by nuclear genes. In total, nearly 1700 nuclear genes are listed as making mitochondrial proteins in the February 2016 release of the MitoProteome database at www.mitoproteome.org, and all but 13 of the mitochondrial proteins—that is, more than 99%—are specified by nuclear genes, being synthesized on cytoplasmic ribosomes and then imported into the mitochondria (see Table 1).
Table1. THE LIMITED AUTONOMY OF THE MITOCHONDRIAL GENOME
The variant mitochondrial genetic code
Prokaryotic genomes and the nuclear genomes of eukaryotes encode many hundreds to usually many thousands of different proteins. They are subject to a “universal” genetic code that is kept invariant: mutations that could potentially change the genetic code are likely to produce at least some critically malfunctional proteins and so are strongly selected against. However, the much smaller mitochondrial genomes make very few polypeptides. As a result, the mitochondrial genetic code has been able to drift by mutation to be slightly different from the universal genetic code.
In the human mitochondrial genetic code, 60 codons specify amino acids, one fewer than in the nuclear genetic code. There are four stop codons: UAA and UAG (also stop codons in the nuclear genetic code) and AGA and AGG (which specify arginine in the nuclear genetic code). UGA, a nuclear stop codon, encodes tryptophan in mitochondria, and AUA specifies methionine in mitochondria, not isoleucine.
There are only 22 different types of human mitochondrial tRNA, but the individual tRNA molecules can each interpret several different codons because of third-base wobble. Eight of the 22 tRNA molecules have anticodons that each recognize families of four codons differing at the third base only. The other 14 tRNAs recognize pairs of codons that are identical at the first two base positions and share either a purine or a pyrimidine at the third base. Between them, therefore, the 22 mitochondrial tRNA molecules can recognize a total of (8 × 4) + (14 × 2) = 60 codons.
