Unexpectedly complex editing patterns at dinucleotide insertion sites in Physarum mitochondria

Abstract

Many of the RNAs transcribed from the mitochondrial genome of Physarum polycephalum are edited by the insertion of nonencoded nucleotides, which are added either singly or as dinucleotides. In addition, at least one mRNA is also subject to substitutional editing in which encoded C residues are changed to U residues posttranscriptionally. We have shown previously that the predominant type of editing in these organelles, the insertion of nonencoded single C residues, occurs cotranscriptionally at the growing end of the RNA chain. However, less is known about the timing of dinucleotide addition, and it has been suggested that these insertions occur at a later stage in RNA maturation. Here we examine the addition of both single nucleotides and dinucleotides into nascent RNAs synthesized in vitro and in vivo. The distribution of added nucleotides within individual cloned cDNAs supports the hypothesis that all insertion sites are processed at the same time relative to transcription. In addition, the patterns of partial editing and misediting observed within these nascent RNAs suggest that separate factors may be required at a subset of dinucleotide insertion sites and raise the possibility that in vivo, nucleotides may be added to RNA and then changed posttranscriptionally.

FIG. 1.

Dinucleotide insertion site contexts. Potential positions of inserted nucleotides are depicted in lowercase type. Note that the insertion positions are uncertain, since all are flanked by encoded residues of the same type as one or both of the extra nucleotides. The extent of this ambiguity is indicated by showing all residues which could be either regularly encoded or added by editing in bold type. The extent of the uncertainty is reduced to the underlined letters if the added nucleotides are inserted as an adjacent pair. The numbering of the editing sites for the ssu gene differs from that of Mahendran et al. (10), as the individual inserted nucleotides at dinucleotide sites are assigned independent numbers, as for the coI gene. LSU, large-subunit rRNA.

FIG. 2.

Patterns of editing and misediting in run-on transcripts synthesized in vitro from the coI gene. A schematic representation of the editing status of sites within individual cDNA clones is depicted, showing 3 dinucleotide insertion sites, 13 sites of C insertion, and 4 sites of C-to-U substitution, as noted. Insertions at dinucleotide editing sites are indicated by the appropriate uppercase letters, while hyphens indicate no insertion. C insertion sites are indicated by symbols as follows: correctly edited, gray diamonds; G misinsertion, black squares; U misinsertion, black circles. The guanosines at es29 and es34 are misinserted downstream of the encoded C nucleotides adjacent to these sites. Lowercase letters show the status of the four substitutional editing sites (the c-to-u sites at es26 to es28 and es30). cDNA clones were generated from run-on transcripts synthesized under either high nucleotide concentrations (500 μM concentrations of each NTP) or low CTP conditions (20 μM CTP, 500 μM ATP, 500 μM GTP, 500 μM UTP). Note that 2 of the 10 sequences in the high-CTP panel are identical; these were obtained in different transformations using the same ligation reaction. All other clones obtained in these experiments are unique.

FIG. 3.

Editing patterns among steady-state (A) and nascent (B) coI transcripts synthesized in vivo. (A) Primer extension sequencing of bulk mitochondrial RNA. Lane —, primer extension in the absence of dideoxynucleotides. (B) Schematic representation of the editing status of sites within individual cDNA clones synthesized from nascent RNAs isolated from mtTECs that had not been subjected to run-on transcription. See text for details. Symbols for C insertion sites are explained in the legend to Fig. 2.

FIG. 4.

Systematic misediting of the AA insertion site (es46/47) in nascent ssu transcripts synthesized in vitro. Patterns of editing at es39 to es47 of the ssu transcript and two downstream C insertion sites (ds1 and ds2) that are present on the same nascent transcript in RNAs isolated from mtTEC before (no run-on; synthesized in vivo) and after run-on transcription in vitro are depicted. Insertions at the dinucleotide site (es46/47) are indicated by the appropriate uppercase letters; -, no insertion; Δ, deletion of the encoded A adjacent to es46/47 (see Fig. 5B for sequence). Symbols for C insertion sites are explained in the legend to Fig. 2. Note that two of the four clones with U inserted at the AA site and the two clones with G inserted at the AA site may not be independent, as they have identical sequences and were generated in the same experiment.

FIG. 5.

Likely sites of dinucleotide insertion based on sequences of partially edited and misedited cDNAs. Alignments of unedited, edited, partially edited, and misedited clones encompassing the GU insertion site of the coI mRNA (es44/45) (A), the AA insertion site of the ssu rRNA (es46/47) (B), and the UA insertion site of the coI mRNA (es31/32) (C). See text for details.