Plant-type RNA editing transferred into Escherichia coli: Expected and surprising new insights for editing factors and how they recognize their targets

Abstract

Plant-type C-to-U RNA editing is a post-transcriptional process that converts cytidines to uridines in mitochondrial and chloroplast transcripts. This process is mediated by pentatricopeptide repeat (PPR) proteins, which recognize specific targets via their PLS repeats and perform cytidine deamination via their C-terminal extension, the DYW domain. In the model moss <em>Physcomitrium patens</em>, all 13 editing sites are fully assigned to 9 DYW-type PPR proteins, making it an ideal model for studying RNA editing mechanisms. Additionally, plant-type RNA editing has been successfully established in other heterologous systems, such as <em>Escherichia coli</em>. The faithful editing efficiency in comparison to plants makes it a powerful system for further investigating RNA editing mechanisms. <br /> The mitochondrial editing factor PPR56 from <em>P. patens</em> efficiently edits its two targets, nad3eU230SL and nad4eU272SL, in both plant and <em>E. coli</em> systems. However, when compared to other PPR proteins such as PPR65 and PPR78, PPR56 exhibits over 100 off-targets in the <em>E. coli</em> transcriptome. Detailed mechanistic studies in the bacterial system have shown that: (i) single amino acid modifications on the PLS motif can redirect PPR proteins to new targets, (ii) the RNA editing activity of PPR proteins is influenced by the target context, and (iii) the editing efficiency is enhanced by tandem targets. A new candidate site, cox3eU290SF, was found to be influenced by PPR56, which suggests a scanning mechanism of PPR proteins along transcripts. <br /> Although C-to-U RNA editing is mainly restricted to land plants, one of the exceptions is the heterolobosean protist <em>Naegleria gruberi</em>, which stores 10 DYW-type PPR proteins in its nuclear genome and has two editing sites in its mitochondrial transcriptome. In this study, NgPPR45 was rearranged, resulting in a DYW-type PPR protein with a mitochondrial signal peptide that nicely fits the two editing sites according to the PPR-RNA binding code, making it an ideal candidate. Since knockout studies could not be performed in <em>N. gruberi</em>, different chimeras of PpPPR78 with NgPPR45 were tested in the <em>ppr78</em> knockout <em>P. patens</em>. The chimera with the PLS stretch of PpPPR78 and the complete C-terminal extensions of NgPPR45 could edit one of the targets for PpPPR78, cox1eU755SL, up to 82%. This proves that the C-terminal extensions of NgPPR45 have a functional cytidine deaminase and can act in fusion with the independently evolved moss PPR stretch. The failure of other chimeras points out that complete C-terminal extensions might be important for fusion proteins from different systems. <br /> PPR proteins, with their target-specific cytidine deamination process, make them an ideal molecular tool for transcriptomic engineering. However, chimeras of PPR proteins do not always work. The PLS stretch of PPR65 is not competitive with the C-terminal extensions of PPR56, while the opposite is true. Changing the DYW domains of PPR56 from different heterologous sources results in restricted or widened off-target numbers in the <em>E. coli</em> transcriptome. The experiments show that the preference of the DYW domains does not influence the immediate sequence of the editing site environment, but rather has a long-range impact on the upstream PPR stretch. <br /> To further investigate the mechanism of RNA editing, PPR proteins can be overexpressed in the cytosol of plants along with a supplied target in knockout plants. Additionally, off-targets in the nuclear transcriptome can be examined to provide a wider range of options for transcriptomic engineering. Due to its high activity and flexibility, PPR56 is an excellent editing factor and could serve as a solid foundation for a molecular tool.