RNA editing refers to changing the sequence of RNA after transcription, either by adding nucleotides, taking them away, or substituting one for another. The extent of editing is dramatic in some mRNAs, e.g. in the mitochondria of trypansomes and Leishmania. For example, for some mRNAs 55% of the nucleotide sequence is added after transcription! In many of the cases characterized so far, a small number of U's are inserted at many places in the mRNA. Other examples of excising U's and adding C's are known for other mitochondrial genes from other organisms.
In at least some cases, the additional nucleotides are added under the direction of guide RNAs that are encoded elsewhere in the mitochondrial genome. A portion of the guide RNA is complementary to the mRNA in the vicinity of the position at which nucleotides will be added (Fig. 3.3.16). The U at the 3' end of the guide RNA initiates a series of phosphoester transfer reactions to insert itself into the mRNA (see bottom of Fig. 3.3.16). More U's at the 3' end of the guide RNA can be added, one at a time. Note the similarity in mechanism between these insertions of nucleotides (editing) and the self‑splicing of Group I intron.
For a situation in which one segment of DNA encodes the unedited mRNA and two other segments of DNA encode the guide RNAs required for editing, the "gene" is encoded in three portions, mutations in which would complement in trans! This is a counter‑example to one of our most powerful definitions of a gene.
In mammals, two different forms of apolipoprotein B are made, one in the liver and one in the intestine. The intestinal form is much shorter because of an earlier termination codon. Surprisingly, only one gene is found and it must encode both from of ApoB. A specifc enzyme must change one nucleotide of the mRNA for apolipoprotein B (a C in codon 2153, CAA) post‑transcriptionally from a C to a U to generate the termination codon (UAA) found in the intestinal form.
This enzymatic activity is present in a protein with no apparent RNA component, and hence no obvious guide RNA. Thus it appears to operate by a distinctly different mechanism from the editing in protist mitochondria (see. e.g. Greeve, J. et al., 1991, Nucleic Acids Research 19: 3569-3576).