C-to-U conversion in the intercistronic ndhI/ndhG RNA of plastids from monocot plants: conventional editing in an unconventional small reading frame?

Editing of plastid RNAs proceeds by C-to-U, in hornwort species also by extensive U-to-C, transitions, which predominantly lead to the restoration of codons for structurally and/or functionally important, conserved amino acid residues. So far, only one instance of editing outside coding regions has been reported - in the psbL/ psbF intergenic region of Ginkgo biloba. This site was proposed to have no functional importance. Here we present an evaluation of an editing site in the ndhI/ ndhG intergenic region in a related group of monocot plants. Efficient editing of this site, as well as the phylogenetic conservation of the resulting uridine residue, point at an important role for the sequence restored by editing. Two potential functions can be envisaged. (1) RNA secondary structure predictions suggest that the C-to-U conversion at this site can lead to a modified stem/loop structure of the ndhG 5' UTR, which could influence ndhG expression. (2) Alternatively, editing of the ndhI/ ndhG intergenic region may tag a so far unidentified small (12-codon) ORF, and lead to the restoration of a conserved phenylalanine codon. A screen with specific antibodies elicited against the putative peptide failed to detect such a peptide in chloroplast fractions. However, this failure may be attributable to its low and/or development-specific expression.     

Pervasive RNA Editing Among Hornwort rbcL Transcripts Except Leiosporoceros

RNA editing affecting chloroplast and mitochondrial genomes has been identified in all major clades of land plants. The frequency of edited sites varies greatly between lineages but hornworts represent an extreme in propensity for editing in both their chloroplast and mitochondrial genomes. cDNA sequences from seven taxonomically diverse hornwort rbcL sequences combined with a survey of 13 additional DNA sequences for potential edited sites demonstrate the presence of 62 edited sites and predict a minimum of 10 additional sites. These 72 total edited sites represent 43 C-to-U and 28 U-to-C nucleotide conversions, with 1 site exhibiting editing in both directions. With one exception, all taxa are heavily edited, with each having from 20 to 34 edited sites. However, a single sample, Leiosporoceros, is shown to lack edited sites. Phylogenetic reconstruction of hornworts results in ambiguous resolution of Leiosporoceros depending on whether edited sites are maintained or eliminated from the analyses. Depending on the inferred relationship of Leiosporoceros to the hornworts, at least two explanations for the origin and maintenance of pervasive editing in hornworts are possible. The absence of edited sites in Leiosporoceros could represent either the absence or a low level of editing ability in the common ancestor of hornworts, as represented by Leiosporoceros, or the loss of editing sites in this lineage after the primary diversification events in the group.