RNA editing in bryophytes and a molecular phylogeny of land plants

RNA editing has been observed to date in all groups of vascular plants, but not in bryophytes. Its occurrence was therefore assumed to correlate with the evolution of tracheophytes. To gain more insight into both the phylogeny of early land plants and the evolution of mitochondrial RNA editing we have investigated a number of vascular and non-vascular plant species. Contrary to the belief that editing is absent from bryophytes, here we report mitochondrial RNA editing in cox3 mRNA of the liverwort Pellia epiphylla, the mosses Tetraphis pellucida and Ceratodon purpureus and the hornwort Anthroceros crispulus. RNA editing in plants consequently predates the evolution of tracheophytes. Editing is also found in the eusporangiate ferns Ophioglossum petiolatum and Angiopteris palmiformis, the whisk fern Tmesipteris elongata and the gnetopsid Ephedra gerardiana, but was not detected in Gnetum gnemon.cox3 mRNA of the lycopsid Isoetes lacustris shows the highest frequency of RNA editing ever observed in a plant, with 39% of all cytidine residues converted to uridines. The frequency of RNA editing correlates with the genomic GC content rather than with the phylogenetic position of a species. Phylogenetic trees derived from the slowly evolving mitochondrial sequences find external support from the assessments of classical systematics.