Trans splicing in Oenothera mitochondria: nad1 mRNAs are edited in exon and trans-splicing group II intron sequences

Identification of editing positions in the ndhB transcript from maize chloroplasts reveals sequence similarities between editing sites of chloroplasts and plant mitochondria

A comparison of the nucleotide sequences from genomic DNA and cDNA of the ndhB gene from maize chloroplasts shows that the ndhB transcript is edited by C-to-U transitions at six positions which appear to exist as editing sites also in the chloroplast ndhB genes from rice and tobacco but not from liverwort. In order to identify possible sequence determinants necessary for editing, the sequences surrounding the newly identified ndhB and previously identified ndhA editing sites were compared with each other and with editing sites observed in plant mitochondrial transcripts. Among the chloroplast editing sites two closely positioned ndhB sites show similarity by sharing a common octanucleotide. The existence of the identical octanucleotide in the ndhJ gene whose transcript is not edited at the respective position, shows, however, that this octanucleotide is not sufficient to elicit the editing process. On the other hand, several of the chloroplast editing sites show sequence similarities with certain sets of consensus sequences reported earlier for editing sites of plant mitochondria. This supports the view that the editing processes of both plant organelles share common components and/or mechanistic steps and that the consensus sequences are part of the determinants necessary for editing.

RNA splicing in lower eukaryotes

Recently, cis-acting elements and trans-acting RNA and protein factors necessary for splicing nuclear pre-mRNAs, group II introns or group III introns, have been discovered, and new roles for the splicing factors have been elucidated. Parallels among the pathways for splicing these different classes of introns have been identified.

RNA editing of atp6 transcripts from male-sterile and normal cytoplasms of rapeseed (Brassica napus L.)

The complete cDNA sequence corresponding to the rapeseed atp6 gene transcript (coding for subunit 6 of F0-ATPase) has been determined by a method involving cDNA synthesis, using specific oligonucleotides as primers, followed by PCR amplification, cloning and sequencing of the amplification products. Only one modification, a C-to-U conversion, has been found when compared to the genomic mitochondrial DNA sequence. Comparison of the extent and frequency of RNA editing of the pol cytoplasmic male sterile (cms) atp6 transcript with those of normal atp6 transcript indicates that there is no variation between the editing status of the atp6 transcripts from pol cms and normal cytoplasms.

Sugarbeet minicircular mitochondrial DNAs: high-resolution transcript mapping, transcript abundance and copy number determination

Three minicircular mitochondrial DNAs have been studied to address several aspects of transcription in sugarbeet mitochondria. High-resolution transcript mapping experiments have shown that sequences at the 5' termini of minicircle transcripts are highly homologous and resemble sequences at the 5' termini of sugarbeet mainband mitochondrial genes (atpA, atp6). In addition, they show homology to transcript termini of mitochondrial genes from other dicotyledonous plants, suggesting they may function as promoter sequences. Conserved sequences, which most probably act as RNA processing signals, were also identified at the 3' termini of minicircle transcripts. An oligonucleotide probe to a 14 base conserved sequence was used to determine the relative copy numbers of the three minicircle components in male-fertile mitochondria. Copy numbers were roughly equivalent, suggesting minicircles are replicated and/or transmitted with nearly equal efficiency, at least in sugarbeet taproots. Mc.a and Mc.c transcript levels are equivalent, consistent with their template copy number, however; Mc.d transcript levels were significantly lower than expected, implicating additional factors such as promoter strength and/or transcript stability in determining transcript levels in sugarbeet mitochondria, as recently demonstrated in maize.

Regenerating good sense: RNA editing and trans splicing in plant mitochondria

The protein products of plant mitochondrial genes cannot be predicted accurately from genomic sequences, since RNA editing modifies almost all mRNA sequences post-transcriptionally. Furthermore, RNA editing alters leader, trailer and intron sequences, and may be required for processing of these sequences. For several plant mitochondrial transcripts, processing includes trans splicing, which connects exons scattered throughout the genome. The mature transcripts are assembled via split group II intron sequences.

Organization of ATPA coding and 3' flanking sequences associated with cytoplasmic male sterility in Phaseolus vulgaris L

A region of the mitochondrial genome associated with cytoplasmic male sterility (CMS) in Phaseolus vulgaris was flanked by two different repeated sequences designated x and y. The DNA sequence of the CMS-unique region and a portion of each flanking repeat was determined. Repeat x contained a complete coding copy of the F1 ATPase subunit A (atpA) gene, as well as an open reading frame (orf) predicting a protein of 209 amino acids. The TGA termination codon of the atpA gene and the ATG initiation codon of orf209 were overlapping. These reading frames were oriented with their 3' ends proximal to the CMS-unique region. The CMS-unique region of 3736 nucleotides contained numerous orfs. The longest of these predicted proteins being of 239, 98 and 97 amino acids. The 3' coding and 3' flanking regions of orf98 were derived from an internal region of the higher plant chloroplast tRNA alanine intron. The region of repeat y immediately adjacent to the CMS-unique region contained the 111 carboxy-terminal coding residues of the apocytochrome b (cob) gene. This segment was oriented with its 5' end proximal to the CMS-unique region, but cob gene sequences were not fused to an initiation codon within the unique region.

Activation of the catalytic core of a group I intron by a remote 3' splice junction

Over 1000 nucleotides may separate the ribozyme core of some group I introns from their 3' splice junctions. Using the sunY intron of bacteriophage T4 as a model system, we have investigated the mechanisms by which proximal splicing events are suppressed in vitro, as well as in vivo. Exon ligation as well as cleavage at the 5' splice site are shown to require long-range pairing between one of the peripheral components of the ribozyme core and some of the nucleotides preceding the authentic 3' splice junction. Consistent with our three-dimensional modeling of the entire sunY ribozyme, we propose that this novel interaction is necessary to drive 5' exon-core transcripts into an active conformation. A requirement for additional stabilizing interactions, either RNA-based or mediated by proteins, appears to be a general feature of group I self-splicing. A role for these interactions in mediating putative alternative splicing events is discussed.

The higher plant nad5 mitochondrial gene: a conserved discontinuous transcription pattern

The single copy nad5 gene has been identified in the mitochondrial genomes of cauliflower, chicory, potato, fennel, and common bean. In these five dicot species the same organization as in Oenothera, Arabidopsis, wheat, and maize has been found for the gene: it consists of five exons organized into three independent groups. The first group comprises exons I and II, separated by a highly conserved group II intron, while the second group consists of exon III only. In the third group exons IV and V are separated by a group II intron of variable, species-specific, length. Transcription analysis of the nad5 gene in chicory and in cauliflower shows that the five exons are assembled as a mature mRNA through intermolecular interactions and multiple splicing events. Comparison of transcription in the gene with that of wheat and maize suggests that a common mechanism exists in higher plants for nad5 transcript processing.

Variable intron content of the NADH dehydrogenase subunit 4 gene of plant mitochondria

The gene nad4, encoding subunit four of the mitochondrial NADH dehydrogenase complex I, has been isolated and characterized from turnip, Brassica campestris. The 8 kb turnip nad4 gene contains four exons, which potentially encode a NAD4 polypeptide of 495 amino acids, and three large group II introns. Northern analysis identifies an abundant 2 kb transcript that most likely serves as the nad4 mRNA, while several larger transcripts (putative splicing intermediates) are also detected. Analysis of the nad4 locus in three distantly related dicotyledons indicates that introns 2 and 3 are optional. Mung bean has the same nad4 organization as turnip, whereas spinach nad4 contains introns 1 and 3, and lettuce nad4 has intron 1 only. We infer that all three group II introns were present in the nad4 gene of an angiosperm common ancestor and have persisted in certain lineages for over 200 million years, with two of the introns having been lost in other lineages.

The coxII gene in carrot mitochondria contains two introns

The gene for cytochrome oxidase subunit II (coxII) in carrot is encoded by a unique locus in the mitochondrial genome. In contrast to the coxII genes in the numerous other plant species investigated to date, the coding region is interrupted by two group II introns. The carrot 5' intron is the homologue of the single intervening sequence found in several monocot and dicot coxII genes. Sequences similar to the 3' intron of the carrot coxII gene have not been reported previously and are not detectable by hybridization with Oenothera mtDNA. Northern hybridizations indicate complex precursor transcript patterns with mRNA molecules up to 10 kb length. The excised intron sequences appear to be stably maintained in the mRNA pool. Amino acid sequence comparisons suggest that the carrot coxII mRNA needs to be edited by numerous C to U transitions.

Large size and complex structure of mitochondrial DNA in two nonflowering land plants

We report the first estimates of genome size and complexity for mitochondrial DNAs (mtDNAs) from nonflowering land plants. The mtDNA of Onoclea sensibilis (sensitive fern) is approximately 300 kb in size, while that of Equisetum arvense (common horsetail) is at least 200 kb. Sufficient mtDNA of Onoclea was available to permit an estimation of the copy number and a linkage analysis of nine mitochondrial genes. Six of these genes appear to be present in only one or two copies in the Onoclea genome, whereas three other genes are present in multiple copies. Five of the approximately ten genes encoding 26S rRNA are located on a large, greater than 10 kb, dispersed repeat that also contains closely linked genes for 18S rRNA and the alpha subunit of ATPase (atpA). The other 26S genes belong to a second dispersed repeat family of greater than 8 kb whose elements do not contain any other identified genes. Because flowering plant mtDNAs are also large and contain dispersed, gene-containing, repeats, it appears that these features arose early in the evolution of land plants, or perhaps even in their green algal ancestors.

Unique patterns of mitochondrial genes, transcripts and proteins in different male-sterile cytoplasms of Daucus carota

Restriction fragment analysis of mitochondrial and chloroplast DNAs from a brown anther and a petaloid cytoplasmic male-sterile (cms) line revealed unique patterns for each cms line distinct from those of normal fertile cytoplasms, but identical restriction fragments for all chloroplast DNAs. The restauration of fertility through the introduction of nuclear restorer genes had no effect on the overall mitochondrial DNA (mtDNA) structure. The genomic environment and transcription patterns of several mitochondrial genes differ between cms and normal cytoplasms, while no difference has so far been detected between cms and the corresponding fertility-restored lines in mitochondrial DNAs and mRNAs. Mitochondrial translation products analysed by in-organello synthesized proteins revealed a number of polypeptides unique to each cytoplasm. Most prominent is a 17-kDa polypeptide that is present in the brown anther cms line but not in fertile mitochondria. Synthesis of this protein was not visibly affected by fertility restauration. The different cms phenotypes in carrot are thus associated with extensive and unique mtDNA rearrangements and distinct alterations in transcription and translation patterns.

Gene organization deduced from the complete sequence of liverwort Marchantia polymorpha mitochondrial DNA. A primitive form of plant mitochondrial genome

Analysis of the mitochondrial DNA of a liverwort Marchantia polymorpha by electron microscopy and restriction endonuclease mapping indicated that the liverwort mitochondrial genome was a single circular molecule of about 184,400 base-pairs. We have determined the complete sequence of the liverwort mitochondrial DNA and detected 94 possible genes in the sequence of 186,608 base-pairs. These included genes for three species of ribosomal RNA, 29 genes for 27 species of transfer RNA and 30 open reading frames (ORFs) for functionally known proteins (16 ribosomal proteins, 3 subunits of H(+)-ATPase, 3 subunits of cytochrome c oxidase, apocytochrome b protein and 7 subunits of NADH ubiquinone oxidoreductase). Three ORFs showed similarity to ORFs of unknown function in the mitochondrial genomes of other organisms. Furthermore, 29 ORFs were predicted as possible genes by using the index of G + C content in first, second and third letters of codons (42.0 +/- 10.9%, 37.0 +/- 13.2% and 26.4 +/- 9.4%, respectively) obtained from the codon usages of identified liverwort genes. To date, 32 introns belonging to either group I or group II intron have been found in the coding regions of 17 genes including ribosomal RNA genes (rrn18 and rrn26), a transfer RNA gene (trnS) and a pseudogene (psi nad7). RNA editing was apparently lacking in liverwort mitochondria since the nucleotide sequences of the liverwort mitochondrial DNA were well-conserved at the DNA level.

RNA editing makes mistakes in plant mitochondria: editing loses sense in transcripts of a rps19 pseudogene and in creating stop codons in coxI and rps3 mRNAs of Oenothera

An intact gene for the ribosomal protein S19 (rps19) is absent from Oenothera mitochondria. The conserved rps19 reading frame found in the mitochondrial genome is interrupted by a termination codon. This rps19 pseudogene is cotranscribed with the downstream rps3 gene and is edited on both sides of the translational stop. Editing, however, changes the amino acid sequence at positions that were well conserved before editing. Other strange editings create translational stops in open reading frames coding for functional proteins. In coxI and rps3 mRNAs CGA codons are edited to UGA stop codons only five and three codons, respectively, downstream to the initiation codon. These aberrant editings in essential open reading frames and in the rps19 pseudogene appear to have been shifted to these positions from other editing sites. These observations suggest a requirement for a continuous evolutionary constraint on the editing specificities in plant mitochondria.

RNA editing in ATPase subunit 6 mRNAs in Oenothera mitochondria. A new termination codon shortens the reading frame by 35 amino acids

The open reading frame encoding ATPase subunit 6 in Oenothera mitochondria is edited at 21 positions in all cDNA clones investigated. Only one of these events is silent, all others improve similarity between the homologous polypeptides of other species. The introduction of a new UAA termination codon shortens the polypeptide by 35 amino acids to a carboxy terminus conserved in other species. In one of the cDNA clones, an additional editing event was observed resulting in a premature UAA termination codon in the amino terminal region.

The mitochondrial genome: so simple yet so complex

Mitochondria possess a small set of genes that are essential for respiratory function. This review highlights recent advances in our understanding of mitochondrial gene organization and expression. These studies illustrate a remarkable diversity among eukaryotic lineages and an impressive complexity of events needed to achieve nuclear-mitochondrial harmony.

Distribution of RNA editing sites in Oenothera mitochondrial mRNAs and rRNAs

To investigate whether RNA editing in plant mitochondria modifies structural RNAs as well as protein-coding RNAs we compared the genomic-encoded information with the respective transcripts of several genes in Oenothera. The genes analysed are the 5S, 18S and 26 S rRNAs, the alpha-subunit of ATPase (atpA), cytochrome b (cytb), orfB, which is located upstream of cytochrome oxidase subunit III, and the respective leader, trailer and spacer sequences. All open reading frames were found to be edited to some degree. The atpA coding region has the least edited mRNA in Oenothera mitochondria, with only four nucleotides altered in the 1533 nucleotide open reading frame. From this analysis we conclude that frequent RNA editing is indicative of functional protein coding regions in plant mitochondria. The extensive editing in orfB, for example, suggests that this orf codes for a mitochondrial protein. No RNA editing event was found in the 5S rRNA or in the 1824 nucleotides analysed of the 18S rRNA, but two nucleotides were found to be altered in the 1970 nucleotides compared for the 26S rRNA. One nucleotide alteration has changed C to U, the other in reverse U to C. However, only one of five cDNA clones covering this region shows the modifications, similar to many silent editing events in open reading frames. RNA editing in the structural RNAs thus does not seem to be essential for their function in the mitochondrial ribosome.

Multiple trans-splicing events are required to produce a mature nad1 transcript in a plant mitochondrion

The mitochondrial gene encoding NADH dehydrogenase subunit 1 (nad1) in Petunia hybrida is split into five exons, a, b, c, d, and e. With the use of a complete restriction map of the 443-kb Petunia mitochondrial genome, we have cloned these exons and mapped their location. Exon a is located 130 kb away from and in the opposite orientation from exons b and c. Exon d maps 95 kb away and in the opposite orientation from exons b and c. Exons d and e are separated by 190 kb. By performing the polymerase chain reaction on Petunia cDNAs, we have shown that transcripts from these five exons are joined via a series of cis- and trans-splicing events to create a mature nad1 transcript. In addition, we have found 23 C----U RNA edit sites in Petunia nad1. RNA editing changes 19 of the amino acids predicted by the genomic sequence.

The wheat mitochondrial gene for subunit I of the NADH dehydrogenase complex: a trans-splicing model for this gene-in-pieces

The nad1 gene encoding subunit I of the respiratory chain NADH dehydrogenase is fragmented into five unique-copy coding segments that are scattered over at least 40 kb and interspersed with other genes in the wheat mitochondrial genome. The nad1 segments are flanked by sequences with group II intron features, and transcript analysis demonstrates the presence of correctly spliced mRNAs. RNA editing occurs at sites asymmetrically distributed along the wheat nad1 coding region, and the initiation codon is created by RNA editing. The unusual organization of the wheat nad1 gene is attributed to mitochondrial DNA rearrangements within introns, and a trans-splicing model involving secondary structural interactions between group II-like intron pieces is proposed for its expression.