Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Available online at www.sciencedirect.com Mitochondrion 8 (2008) 35–46 www.elsevier.com/locate/mito The process of RNA editing in plant mitochondria Mizuki Takenaka *, Daniil Verbitskiy, Johannes A. van der Merwe, Anja Zehrmann, Axel Brennicke Molekulare Botanik, Universität Ulm, 89069 Ulm, Germany Received 15 August 2007; received in revised form 24 September 2007; accepted 26 September 2007 Available online 11 October 2007 Abstract RNA editing changes more than 400 cytidines to uridines in the mRNAs of mitochondria in flowering plants. In other plants such as ferns and mosses, RNA editing reactions changing C to U and U to C are observed at almost equal frequencies. Development of transfection systems with isolated mitochondria and of in vitro systems with extracts from mitochondria has considerably improved our understanding of the recognition of specific editing sites in the last few years. These assays have also yielded information about the biochemical parameters, but the enzymes involved have not yet been identified. Here we summarize our present understanding of the process of RNA editing in flowering plant mitochondria. Ó 2007 Elsevier B.V. and Mitochondria Research Society. All rights reserved. Keywords: Plant mitochondria; RNA editing; Cytidine deaminase; RNA helicase 1. Introduction In mitochondria and plastids of all flowering plants examined, the sequences of many transcripts are altered by numerous post-transcriptional conversions of C to U (Fig. 1; Covello and Gray, 1989; Gualberto et al., 1989; Hiesel et al., 1989). In plastids, including chloroplasts of flowering plants, in total only about 35 Cs are found to be deaminated to Us, details of which have recently been summarized (Shinakai, 2006). In Arabidopsis thaliana, Brassica napus and Oryza sativa, the entire sets of editing sites in the protein-coding regions of all identified mRNAs have been investigated. These amount to 441, 427 and 491 editing sites per mitochondrial transcriptome in these species, respectively (Giegé and Brennicke, 1999; Notsu et al., 2002; Handa, 2003). The total number of C to U changes will be somewhat higher, since several events have been documented in non-protein-coding regions such as tRNAs and introns, as well as * Corresponding author. Tel.: +49 731 5022677; fax: +49 731 5022626. E-mail address: [email protected] (M. Takenaka). in leader and trailer regions, but not all of these regions have been analyzed in any plant species. These are comparatively few editing sites, since by extrapolation, only about 20–50 are expected in a given transcriptome. RNA editing is required for gene expression in plant mitochondria. The genomic information encoding an open reading frame or a tRNA is more often than not incomplete in these organelles and will not yield a functional product. Therefore, gene expression in plant mitochondria depends on RNA editing for its biological functionality. In all flowering plant mitochondrial systems, RNA editing is essential for the synthesis of correct proteins which, after editing, generally exhibit closer sequence conservation with their homologs in other systems. In addition, RNA editing in plant mitochondria is required in several tRNAs to be able to fold correctly and thus to become functional (Marchfelder and Binder, 2004). RNA editing in plant mitochondria is an essential step of RNA maturation without which no working respiratory chain and no functional mitochondria can be assembled and maintained in the cell. In this review, we will first summarize the observed features of RNA editing in plant mitochondria in the steadystate RNA population and their functional consequences. 1567-7249/$ - see front matter Ó 2007 Elsevier B.V. and Mitochondria Research Society. All rights reserved. doi:10.1016/j.mito.2007.09.004 36 M. Takenaka et al. / Mitochondrion 8 (2008) 35–46 Fig. 1. RNA editing in mitochondria of flowering plants is a posttranscriptional process which changes selected cytidines (C) in the primary transcript to uridines (U) in the mature mRNA. The amino acid sequence encoded by the fully edited, mature mRNA is different from the protein sequence encoded by the genomic DNA and the primary transcript. In this hypothetical sequence several examples of such amino acid changes are included. About 400–500 such C to U RNA editing events are observed in the mRNA and tRNA population in plant mitochondria of a given angiosperm species. Since only some of the cytidines are altered by RNA editing, these have to be specified and distinguished from cytidines which are not changed. In the second part, we will look at recent advances in the analysis of the parameters of site recognition and of the biochemistry involved. 2. How was RNA editing established in the plant kingdom? In flowering plant mitochondria RNA editing involves exclusively C to U changes, in which the C in the premRNA is presumably deaminated to a U moiety (Fig. 2). Most RNA editing events are found in the coding regions of mRNAs and less frequently in introns and other nontranslated regions. RNA editing in tRNA molecules can restore essential base-pairings and only complete RNA Fig. 2. The biochemical reaction of RNA editing in plant mitochondria is in its effect a deamination step. The amino group from the base of the cytidine may be removed by a modified cytidine deaminase like enzyme or may be transferred by a modified transaminase to a receptor molecule. This latter should be a larger moiety such as another RNA molecule, since in the various functionally competent in vitro assays small molecules have been removed and thus do not appear to be required. editing will then allow correct folding and processing of the tRNA precursors. In the ribosomal RNAs in plant mitochondria as well as in chloroplasts no RNA editing events have been observed so far and it is not clear why these RNAs are exempt from editing. In flowering plant species, U to C conversions do not occur or are very rare events. In the plants A. thaliana, B. napus and O. sativa, none have been identified (Giegé and Brennicke, 1999; Notsu et al., 2002; Handa, 2003). In non-flowering plants, the amination of U to C is much more common. In hornworts, for example, this reaction almost reaches the frequency of the C to U editing events (Malek et al., 1996; Knoop, 2004). The only land plants in which no RNA editing have been observed is the liverwort Marchantia polymorpha (Oda et al., 1992) and species of the Marchantiales closely related to M. polymorpha (Malek et al., 1996). Liverworts of other orders, for example, the Metzgeriales, do show frequent RNA editing at least in their mitochondrial mRNAs (Malek et al., 1996). Interestingly none of the closest relatives to the land plants in the group of the green algae, such as the Charales, show any RNA editing (Turmel et al., 2003). From these observations it has been postulated that the establishment of RNA editing in plant mitochondria may have occurred at some point of the early evolution of land plants, the Marchantiales presently being considered as branching very near or at the very root of the land plant tree (Malek et al., 1996; Knoop, 2004). The presence of frequent U to C aminations among the mosses and ferns, for example in hornwort species and in the Lycopodiopsida (Knoop, 2004; Groth-Malonek and Knoop, 2005) has consequences for the biochemical reactions and therefore the enzymes possibly involved. The characteristics of the previously characterized cytidine deaminases and cytidine synthases make it unlikely that a single enzyme of either of the two classes can perform both reactions. Details of these groups of enzymes are discussed further below. The asymmetric distribution of editing sites between coding and non-coding regions in RNAs suggests that there is some selective pressure at work. Without a functional necessity to maintain an editing site, this would soon be lost. The requirement to produce proteins with certain amino acid sequences may be this selective pressure in the open reading frames of mRNAs, where only a U at a given editing position will yield a specific codon identity. In introns, RNA editing is sometimes a prerequisite of correct folding of essential secondary structures and, thus, efficient intron excision (Carillo and Bonen, 1997), which latter is of course required to produce a functional protein product. This requirement for RNA editing in plant mitochondria has most likely evolved after the biochemical potential to post-transcriptionally modify certain nucleotide positions was established. This order of events was first formulated by Covello and Gray (1993), where the authors suggest that first a deaminating/aminating enzyme had M. Takenaka et al. / Mitochondrion 8 (2008) 35–46 mutated from acting only on mononucleotides or whatever other substrates to also accept polynucleotide templates. In the second step, the mitochondrial DNA was then allowed to mutate by changing a T to a C (or a C to a T) in those positions that could be addressed and corrected post-transcriptionally by the novel RNA editing enzyme. Once the first such mutation requiring this compensating correction in the RNA had occurred, RNA editing was established and had to be maintained. This order of events was aptly named the bureaucracy model (Mulligan, 2004), because of its similarity to any administration which will introduce often unnecessary complexities with the sole aim of making itself indispensable. 3. What can RNA editing in plant mitochondria be good for? In this bureaucracy model RNA editing is just a waste of time and energy, because, without exception, all of the RNA editing events described to date would be rendered unnecessary if the sequence of the mature mRNA were encoded in the genome, as is the case for example in bacteria. If there is no obvious advantage in the existence of RNA editing, there may be a hidden biological importance of RNA editing. One possibility of such a camouflaged agenda might be another level of regulatory control of gene expression among the various post-transcriptional processes. For example, the introduction of an AUG translational start codon from an ACG codon could make an mRNA rapidly accessible for translation. This would be much faster than the complete de novo synthesis of the affected transcript (Fig. 1). RNA editing could confer an advantage with its potential of generating mRNA variants, which will allow to synthesize two or more different polypeptide products from a single gene. This potential is realized by some viral editing systems and most notably in the tissue-specific apolipoprotein B editing in mammals. The site specific C to U deamination changing a CAA codon in the apolipoprotein B mRNA to a UAA stop codon is controlled tissue specifically. In different cells two distinct lipid carrier proteins of 100 kDa and 49 kDa, respectively, are synthesized from one and the same genomic coding region (Greeve et al., 1991; Chang et al., 1998). On a larger scale, the A to I RNA editing in the GluR receptor channels results in a large array of variant proteins available for the combinatory assembly of these channels which are involved in fine-tuning the properties of the chemo-electric potential transmission (O’Connell, 1997; Sommer et al., 1991). This latter option of generating polypeptide variants should be reflected in the access of the translation system to the partially edited mRNAs observed in plant mitochondria for some mRNA species. Indeed, experimental examinations showed that partially edited mRNAs are present in polysomal fractions of plant mitochondria and can apparently be translated into a family of variant proteins (Mulligan, 2004). However, these variant proteins are most 37 likely non-functional and get rapidly degraded, since only one type of protein sequence appears to be incorporated into the polypeptide complexes of the respiratory chain (Grohmann et al., 1994). These integrated polypeptide sequences correspond to the protein specified by the fully edited mRNA with the omission of the rare sporadic editing events which are most likely errors of the editing machinery (Grohmann et al., 1994; Phreaner et al., 1996; Mulligan, 2004). In fact, it is unlikely that polypeptides synthesized from unedited RNAs would function properly. These variant proteins would rather impair the efficiency of mitochondrial respiration, since the polypeptide sequences derived from the fully edited mRNAs are generally better conserved with the homologs in other organisms than the proteins encoded by the genomic sequences without editing or by any of the editing intermediates. Therefore, the proteins encoded by the fully edited mRNAs have been extensively selected for their physiological and biochemical functionality. The possible functionality of proteins made from unedited mRNAs has been experimentally investigated for the atp9 gene. The (unedited) mitochondrial genomic sequence was integrated into the nuclear genome of tobacco with the proper expression signals as well as with correct mitochondrial import addresses fused to the open reading frame. The synthesized and imported ATP9 protein variant is not functional since its presence disturbs mitochondrial function, presumably by competing with the endogenous native, mitochondrially encoded protein. The resulting phenotype manifests itself in a dysfunction of the pollen development and results in male sterility (Hernould et al., 1993; Zabaleta et al., 1996). The large size of the plant mitochondrial genomes with their ample space to encode additional genes or their variants make it unlikely that RNA editing would be advantageous to increase the coding capacity of a given mitochondrial gene just to save space. Furthermore, such a presumed advantage would be illusionary because the RNA editing process itself requires a number of additional polypeptides. Thus we are presently left with the bureaucracy model of RNA editing being a waste of time and energy, with RNA editing being a biologically selfish process, which once it has become established now has to be perpetuated. However, there is still a third possible raison d’etre, a long term evolutionary advantage. This may be to prevent or at least to complicate transfer of mitochondrial genes to the nuclear genome. The mitochondrion may profit from the establishment of RNA editing as a means to protect its genome. Since there is no evidence of editing in the nucleus/cytoplasm of plants, a gene sequence requiring editing in the mitochondrion could not immediately become functional upon integration in the nuclear genome. A successful transfer would be most easily achieved by reverse transcribing the respective coding region from a fully edited mitochondrial mRNA molecule. The resulting 38 M. Takenaka et al. / Mitochondrion 8 (2008) 35–46 cDNA sequence would encode the correct protein sequence upon appropriate activation in the nuclear genome. However, the reverse transcription step is not straightforward and so far mostly unedited genomic mitochondrial sequences have been found to be inserted into the nuclear genomes of various plants (Bergthorsson et al., 2003; Blanchard and Schmidt, 1995; Brennicke et al., 1993; Stupar et al., 2001). The protective effect of the editing requirement may be real since mitochondria in many other species protect their genes from transfer by using variants of the codon translation system. In these organisms (such as mammals) genes from the mitochondrion cannot be directly translated into functional proteins in the nucleus/cytoplasm since several essential amino acids will be decoded wrongly. The coincidence of identical decoding systems with the same translational cipher in plant mitochondria (and plastids) and in the cytoplasmic system is probably unrelated to the advent of RNA editing, but may be a fortuitous consequence of the established import of various tRNA species from the cytoplasm into mitochondria. Since the mitochondrion and the cytoplasm share several of these tRNAs, they will impose a stringent selection pressure to maintain the standard code in both compartments. 4. How is an RNA editing site identified in plant mitochondria? As the name implies, RNA editing in plant mitochondria and in chloroplasts is a post-transcriptional process. The identification of partially edited mRNA molecules was the first hint that RNA editing is a process separated from transcription and that the nucleotide changes are not introduced by canonic signals in the genomic DNA. In such partially edited transcripts, some of the Cs at potential editing sites have been altered into Us while some the Cs encoded by the genome, but found edited to Us in most mRNA molecules, are (still) unedited. Subsequently the successful in vitro assays in which respective changes were observed in added RNA molecules clearly confirmed this conclusion. These partially edited RNA molecules furthermore suggest that each editing site is identified individually with some being more rapidly edited than others. To differentiate a C to be edited from a C-moiety not to be altered, these need to be discriminated. The respective information to identify a given editing site can only be encoded in its specific nucleotide sequence context. Theoretical calculations taking into account the genome size and the number of RNA editing sites suggest that ten nucleotides would be sufficient to identify a nucleotide sequence motif unambiguously in a plant mitochondrial genome. In the past few years detailed studies of the parameters selecting RNA editing sites within an RNA molecule by in organello and in vitro assays have yielded considerable advances in our understanding of this process. In the following we summarize these insights gathered in various plant species. The very similar observations made in the different flowering plants suggest that one and the same recognition process has been maintained through the evolution of the higher plants. 5. In vivo information on the specificity determinants Genome internal recombinations in the monocot plant rice and in the dicot plant Oenothera created duplicated fragment of genes. From these fragments transcripts are generated in vivo in which some of the editing sites of the respective original open reading frame are present but have retained only few of the nucleotides surrounding this editing site in the original context (Lippok et al., 1994; Kubo and Kadowaki, 1997). In those instances where upstream sequences were retained, these duplicated sites were seen edited. Conversely, when only the downstream context was duplicated, no editing was observed. These natural chimeric sequences had thus already indicated that most of the specificity determinants must be located upstream and that downstream as little as three conserved nucleotides could be sufficient to identify a given editing site (Lippok et al., 1994). In silico comparisons of the sequences surrounding editing sites, e.g. all the 441 sites in the protein-coding regions of the A. thaliana mitochondrial RNA, did not reveal common sequence motif(s) in all, many or even only several editing sites (Giegé and Brennicke, 1999). The only common feature observed in the vicinities of editing sites is preferences for the nucleotide identity immediately adjacent to the edited C. Most notably are guanosines rarely observed immediately upstream of the edited C. The editing enzyme or complex thus seems to sterically prefer pyrimidines at this position while a guanosine may make access more difficult. In chloroplasts, in vivo analysis by transformation has been established for tobacco and has yielded important insights into the structure and extension of cis-elements in this organelle (Chaudhuri et al., 1995; Chaudhuri and Maliga, 1996; Bock et al., 1996, 1997; Reed et al., 2001; Chateigner-Boutin and Hanson, 2002). This approach is so far not possible for plant mitochondria since no successful transformation has as yet been reported mostly due to the lack of a selectable marker gene specific for mitochondria. 6. Experimental investigations of the specificity determinants The requirements of size and location of the specificityconferring nucleotide context relative to the edited nucleotide in plant mitochondria have recently been experimentally investigated in detail by in organello and in vitro approaches. The successful establishment of in organello editing systems for wheat and maize (Farré and Araya, 2001) allowed the characterization of the cis-requirements at individual sites in these two monocot plant species. In vitro RNA edit- M. Takenaka et al. / Mitochondrion 8 (2008) 35–46 ing systems were first developed for chloroplasts (Hirose and Sugiura, 2001; Miyamoto et al., 2002, 2003). Analogous to these, reliable in vitro activities could be established for pea and cauliflower mitochondria (Takenaka and Brennicke, 2003). 6.1. In organello RNA editing in wheat and maize The first in organello assay was successfully established by the group of Alejandro Araya for wheat mitochondria (Farré and Araya, 2001; Choury and Araya, 2006). The procedure was successfully adapted to maize mitochondria in the laboratory of Frank Kempken (Staudinger and Kempken, 2003). Briefly, DNA molecules coding appropriate promoter and RNA stability structures and non-mitochondrial sequences for discrimination from the endogenous mRNAs are introduced into intact isolated mitochondria by electroporation. The DNA is transcribed in the organelles, and the RNA is processed by correct intron excision and RNA editing at a number of the bona fide in vivo RNA editing sites. Several of these actively edited sites have been investigated in detail by various mutations of the nucleotide sequences surrounding the analyzed editing site. In the two monocot plants wheat and maize about 16 nucleotides in the upstream region and at most six nucleotides downstream are sufficient to determine at least some of the nucleotide positions to be edited (Farré et al., 2001; Choury et al., 2004). For example, the extensively analyzed editing sites 77 and 259 in the cox2 mRNA are correctly recognized in wheat mitochondria when this window of surrounding nucleotides is present. More distant nucleotides have little influence, since relocation of these 23 respective nucleotides into a different further sequence context in the mRNA still confers correct RNA editing at this C (Choury et al., 2004). The sequence window requirement of 23 nucleotides is similar for these two editing sites in the cox2 mRNA. The observation that the importance of individual nucleotide identities (such as the nucleotide immediately 3 0 of the edited C) differ between the two sites, suggests that editing sites in plant mitochondria are identified by individual and specific trans-recognition factors in the assembly of the hypothetical editosome. In maize mitochondria the in organello editing system shows that, in some instances, more than the immediate vicinity of an editing site may be required for efficient editing (Staudinger and Kempken, 2003; Staudinger et al., 2005). While transcripts synthesized in the purified mitochondria from an introduced cox2 gene from the rather distant dicot A. thaliana were edited at several sites, an atp6 mRNA from the more closely related plant Sorghum bicolor was not edited at all, although in these transcripts all editing sites are identical with maize and their surrounding sequences are also largely conserved (Bolle and Kempken, 2006). These results suggest that variations in distant 39 regions, possibly even the structures of 5 0 - and/or 3 0 untranslated sequences, might influence RNA editing. 6.2. In vitro RNA editing in pea and cauliflower The in vitro RNA editing systems from pea and cauliflower mitochondria overall yielded similar results on the recognition of RNA editing sites in mRNAs of the two dicots as in the in organello systems of the two monocots. This suggests that the basic parameters of the RNA editing machinery have been conserved in flowering plants. In both dicot plants about 20 nucleotides are essential upstream, 40 are optimal, and none are necessary downstream of the first RNA editing site in the atp9 mRNA. Analysis of the cisrequirements by targeted mutations of the template and competition experiments narrowed the sequence requirements for the site specificity to the region 5–20 nucleotides upstream of this atp9 mRNA site (Takenaka et al., 2004). The region up to 40 nucleotides upstream influences the efficiency of in vitro editing (Neuwirt et al., 2005). If this sequence region is deleted, in vitro editing is still observed, but with less than 50% of the activity recorded with the full-length template. This region surprisingly also influences editing at the next downstream site, which is 30 nucleotides 3 0 of the first site, and thus acts across a total distance of about 70 nucleotides (van der Merwe et al., 2006). Here species-specific adaptations are observed, since the native cauliflower 20 to 40 sequence region enhances the reaction, while the heterologous pea sequence inhibits editing in the cauliflower lysate. This long-distance effect is analogous to the observations made with a heterologous sorghum atp6 sequence in electroporated maize mitochondria (Staudinger et al., 2005). A homologous maize sequence is required far upstream of the conserved atp9 reading frame of sorghum origin to confer correct and efficient RNA editing in organello. These observations suggest that there is a species-specific RNA editing entry motif in this upstream region. The results furthermore indirectly suggest that the RNA editing factor(s) move in some mode of linear progression along the mRNA molecule. However, only few and apparently randomly distributed conserved nucleotide sequence motifs have been found upstream of open reading frames in plant mitochondria (Schuster and Brennicke, 1989; Pring et al., 1992), and it is questionable whether these might play a role in such a function. 6.3. Long distance sequence effects How can the observed long distance sequence effects be explained mechanistically? A straightforward explanation would involve the editosome directly and would postulate some type of scanning progression of the editing complex. This model would, however contradict the observed individual and independent addressing of each editing site through random recognition of the surrounding specific sequence motifs. 40 M. Takenaka et al. / Mitochondrion 8 (2008) 35–46 An alternative, more indirect explanation might be the engagement of RNA helicases and potential cofactors, which are required to clear the template RNA for access of the editosome (Fairman et al., 2004). A sequence motif active in one species may not be sufficient, or may be a hindrance, for a helicase (complex) from another plant species. The resulting mode of action of the RNA editing process would appear as an intermediate between a scanning and a hit-and-run mechanism, and would show characteristics of either. Depending on the experimental assay, the scanning of the RNA helicase would appear as dominant characteristic or the apparently random editing of individual sites such as manifested in the appearance of partially edited sites in the in vivo mRNA population would be the prominent feature observed. 6.4. Site recognition The hit-and-run model, according to which specific regions of the RNA molecule are addressed individually and randomly by the RNA editing machinery, was tested by detailed in vitro experiments (Verbitskiy et al., 2006). This investigation showed that partially edited transcripts are true intermediates, which can serve as substrates for further rounds of RNA editing. It is particularly striking that even substrate RNA molecules, in which one of two closely spaced editing sites was already modified, are accepted as substrates for further rounds of RNA editing. In fact, editing efficiencies in such partially edited RNAs are comparable to those in completely unedited RNA molecules. This result shows that recognition of a specific editing site by the respective trans-factors is independent of the nucleotide identity at the actual editing site. Site recognition and site specificity generally appear to occur at random contact points along the mRNA molecule. The vicinity of each site carries its own recognition context, resulting in an apparent hit-and-run process of RNA editing. Most notably in this respect are the observed efficiencies of competition experiments of in vitro assays and the results of the transfection assays with isolated wheat mitochondria (Choury et al., 2004; Verbitskiy et al., 2006). These in vitro and in organello competition assays show that a homologous sequence can compete RNA editing at a given site, while no competition with the sequence contexts of different editing sites is observed. These competition assays further underline the importance of the specific upstream elements. A wild type upstream sequence can compete as long as it is competent as a template. The competition effect is lost for mutated sequences which are not edited in vitro since they no longer contain the individual site-specific nucleotide positions essential for recognition. Position and extent of these essential nucleotides upstream (and at very few nucleotides downstream) have been found to vary between different editing sites. This observation suggests that not only the variant primary sequences but also the extent of the contact points of the specific trans-factors can vary between indi- vidual recognition elements and is crucial for efficient editing. The experimental results from in vitro assays with dicots and in organello investigations with monocots furthermore suggest that while the recognition parameters can vary between individual sites, the characteristics of the recognition of a given RNA editing site may have maintained similar features of primary sequence and extension between different plant species. Since most of the 400 RNA editing sites in flowering plant mitochondria are conserved between different species and since most of these sites are in conserved mRNAs or tRNAs, the sequence vicinities of homologous sites are also very similar. Consequently, it can be postulated that in different plants these sites are recognized by likewise conserved trans-factors. Therefore, to identify an individual conserved RNA editing site, distinct trans-factors have to be involved which may have been conserved to some extent during the evolution of the flowering plants. Consequently, several hundred distinct, but at least partially conserved trans-factors have to be postulated to address each of the 400–500 editing events in plants mitochondria. 6.5. Recognition of RNA editing sites by protein-factors? To correctly and unambiguously identify this large number of editing sites in mitochondrial transcripts, a correspondingly large number of trans-factors will be required. Since the search for guide-RNA like molecules has been unsuccessful so far, sequence-specific recognition may be mediated by proteins alone as trans-factors. The number of different site-specific proteins required to specify all of the sites will require several hundreds of corresponding genes in the nuclear genome of plants. Recently, the protein (and corresponding gene) family of the so-called PPR proteins has been proposed as a candidate group for these specificity factors. Its name derives from the common presence of numerous pentatricopeptide repeats, which vary in number and in their combinations with other domains in various subgroups. In plants, this family of proteins is highly amplified and contains several hundred members (Small and Peeters, 2000; Lurin et al., 2004). The profiles of prediction programs suggest that a large proportion of the encoded proteins is indeed targeted to either plastids or mitochondria, or both (Lurin et al., 2004). Experimental evidence for several members with full-length reading frames or with their N-terminal regions in transgenic plants or in transient intracellular assays confirms that these are competent to target the respective proteins to these organelles. The first theoretical considerations that these proteins could be nucleic acid binding are confirmed by genetic and biochemical evidence, which has shown that several of these PPR proteins are involved in various processing steps of RNAs in plastids. For these functions in intron excision and endonucleolytic processing binding of the RNA substrate is essential. M. Takenaka et al. / Mitochondrion 8 (2008) 35–46 The first clear connection between the PPR proteins and RNA editing was also made in the chloroplast, where inactivation of a specific PPR protein encoding gene in A. thaliana leads to the loss of one specific RNA editing site (Kotera et al., 2005). A second example of such a connection has recently been found between a second such PPR protein and another editing site in plastids (Okuda et al., 2007). The family of PPR proteins is thus presently the favored candidate to contribute the site-specific recognition factors for RNA editing in plastids as well as in mitochondria. Recently genetic approaches have been initiated to identify the trans-acting specificity factors in A. thaliana (Bentolila et al., 2005; Anja Zehrmann, Johannes A. van der Merwe, Mizuki Takenaka, unpublished results). These assays take advantage of ecotype-specific variations in the RNA editing efficiency at some of the sites. Several essential sites have been identified beyond the one previously reported (Bentolila et al., 2005), which are in one ecotype edited completely or nearly completely in the in vivo mRNA population, but are edited to only about 40–60% in another ecotype (Anja Zehrmann, Johannes A. van der Merwe, Mizuki Takenaka, unpublished results). These QTL-traits are presently being followed through their pattern of inheritance and true nuclear inheritance may eventually lead to the identification of the loci responsible and may identify one or more of the genes encoding these specificity-mediating proteins in flowering plants. 7. Which enzymes may do the RNA editing in plant mitochondria? In plant mitochondria (and plastids) no nucleotide insertions or other changes of nucleotide identities, such as those identified in kinetoplasts (Benne et al., 1986; Chang et al., 1998; Hajduk and Sabatini, 1998) or in mitochondria of the slime mould Physarum polycephalum (Mahendran et al., 1991; Gott and Visomirski-Robic, 1998) have been observed. Therefore, the most parsimonious explanation of the enzymatic mechanism of RNA editing in plants would be to postulate the involvement of an RNA-specific C-deaminase analogous to the APOBEC-1 enzyme described in the apolipoprotein B mRNA editing of mammals (Greeve et al., 1991; Navaratnam et al., 1998). Several lines of experimental evidence indeed suggest that a deamination reaction may be performing the C to U conversions in plant mitochondria (Araya et al., 1992; Rajasekhar and Mulligan, 1993; Yu and Schuster, 1995; Mulligan, 2004). Assays with in organello and in vitro systems have investigated the fate of the sugar-phosphate backbone as well as the nucleotide base (Rajasekhar and Mulligan, 1993; Yu and Schuster, 1995). These experiments have shown that the sugar-phosphate backbone and the nucleotide base are retained intact during and after the editing process, which excludes nucleotide excision and base exchange as possible reactions. These results leave deamination or 41 transamination to mediate the resulting C to U change in flowering plant mitochondria (Fig. 2). 7.1. Are transamination reactions involved? Transamination reactions transferring an amino group from one molecule to another occur in many of the various pathways of amino acid biosynthesis. Usually these reactions result in the amination or deamination of an intermediate of the compound to be synthesized and often require another compound as acceptor of the amino group. In the in vitro RNA editing system we have tested several of these acceptor molecules such as oxaloacetate and a-ketoglutarate for their effect on the editing reaction (Mizuki Takenaka, unpublished data). None of these low molecular weight molecules showed any influence on the efficiency of the in vitro editing reaction. In addition, these small molecules such as oxaloacetate and a-ketoglutarate with molecular weights below 10 kDa should be quantitatively removed from the in vitro plant mitochondrial lysates during the extensive dialysis step of the extract. Since the RNA editing reaction proceeds in vitro without their addition to the assay, these or similar compounds should not be required as amino group receptors. However, if these enzymes had evolved to accept as substrates polynuclotides such as a long RNA molecule they may also have evolved to accept different amino group receptors, for example longer molecules such as another RNA. Therefore such a transamination reaction cannot be excluded per se to be involved in the C to U deamination observed in flowering plant mitochondria. One such candidate enzyme is the enzyme glutamate dehydrogenase (GDH). GDH can catalyze amination as well as deamination reactions depending on the presence and concentration of various allosteric and isosteric regulators (Fig. 3). There is thus a (albeit remote) possibility that one or more of these enzymes have been recruited into the RNA editing reaction to catalyze the deamination step from C to U. Surprisingly representatives of the glutamate dehydrogenase group of proteins, i.e. GDH1 and GDH2, were found among a number of general plant mitochon- Fig. 3. The household enzyme glutamate dehydrogenase catalyzes reactions which include deamination and amination steps. The compounds NAD(P) or NAD(P)H are cofactors for the GDH enzymes and are required for the respective directions of the activity. The equilibrium of the reaction is allosterically influenced by the ambient concentrations of ATP. One of the GDH proteins or a similar enzyme could be a candidate for the evolutionary origin of the plant mitochondrial and plastid RNA editing enzyme. Mutations could have altered such an enzyme to be able to contact RNA molecules and to deaminate specific cytidines and in addition to aminate certain uridines. 42 M. Takenaka et al. / Mitochondrion 8 (2008) 35–46 drial proteins identified by binding to RNA editing templates (Takenaka et al., 2007). The potential participation of the GDH in the C to U deamination reaction was investigated by testing the influence of various cofactors of this enzyme on the in vitro editing assay. The most prominent and essential cofactors of the normal GDH catalyzed reactions are NADH and NADP for the deamination and amination reactions, respectively. However, NADH as well as NADPH stimulated the reaction while the oxidized dinucleotides NAD and NADP did not. To resolve this contradiction and to determine whether the GDH is indeed required for the RNA editing reaction, the effect of a direct blocker of the GDH proteins, the compound phosphinothricin, was analyzed. Surprisingly addition of phosphinothricin to the in vitro RNA editing reaction did not inhibit, but rather stimulated the in vitro editing activity (Fig. 4; Takenaka et al., 2007). This result clearly shows that the GDH polypeptides are not the active enzymes involved in RNA editing in plant mitochondria. 7.2. Is a cytidine deaminase involved? These recent results and arguments make the participation of a modified cytidine deaminase as the responsible enzyme more likely. A family of nine cytidine deaminase genes has been described in the nuclear genome of A. thaliana (Faivre-Nitschke et al., 1999). One or more of these enzyme homologs may be targeted to mitochondria (and/ or plastids) and could thus be involved in the RNA editing reactions in mitochondria (and plastids), leaving others available for the classic role of cytidine deaminase in the Fig. 4. The enzyme glutamate dehydrogenase (GDH) is not involved in RNA editing in plant mitochondria. This is evidenced by experimental results, which show that the GDH-specific inhibitor phosphinothricin does not inhibit in vitro editing. On the contrary, GDH supresses in vitro editing, since phosphinothricin substitutes the requirement for NTP in the in vitro RNA editing reaction. In this experiment, RNA editing of the first editing site in an atp4 template RNA is observed in the presence of 15 mM CTP (left lane; the fluorescent cDNA molecules are derived from TDGrecognized mismatches due to the C to U editing reaction in the template RNA; Takenaka and Brennicke, 2003). In the control in vitro incubation without any added nucleotide no editing signal is detected (lane marked 0). In the presence of 10 or 15 mM phosphinothricin (lanes labeled ‘‘phosphinothricin’’) the products of in vitro editing reactions are detected although no NTP or dNTP had been added. Only the relevant portion of the gel image is shown. nucleotide biosynthesis pathway from cytidine towards uridine. Presently there are two circumstantial arguments against the participation of such a classic type of cytidine deaminase. The first difficulty derives from the experimental failure to inhibit the in vitro RNA editing activity by extraction of the usually required and usually accessible zinc ions from the reaction. All classic cytidine deaminases contain reaction-centre bound zinc and can be readily incapacitated by specific zinc chelators such as 1,10 o-phenanthroline. The bacterial cytidine deaminases as well as the apolipoprotein B cytidine deaminase are both completely blocked by addition of this chelator (Carter, 1998). Similarly, in chloroplast in vitro assays, the zinc chelator has some detrimental effect on the editing reaction (Hegemann et al., 2005). The in vitro plant mitochondrial RNA editing reaction is however not inhibited by the addition of this compound (Takenaka and Brennicke, 2003). The plant mitochondrial enzyme thus would have to have evolved differently to protect the central zinc ion. Indeed, other enzymes have been identified, in which the essential zinc atoms are bound so tightly inside the folded protein that none of the chelators tested can completely remove these metal atoms (Späth et al., 2007). Alternatively, a different C-deaminating enzyme may not require a bound zinc atom. The second, less direct argument against the involvement of a classic cytidine deaminases derives from the occurrence of the ‘reverse’ editing events of U to C so frequently observed in ferns and mosses. For energetic considerations, cytidine deaminases are unlikely to be able to catalyze this reaction and add an amino group to a U residue to generate a C. Thus, a second unrelated enzyme would have to be invoked for these reverse reactions. However, the more parsimonious solution would be a single enzyme able to perform both reactions. Slightly diverging evolutionary paths may have lead to an enzyme in hornworts and related plant species which is able to catalyze the U to C reaction, while the variant protein in the flowering plants would be less suited to facilitate this reaction. On the other hand parallel evolution of various enzymes involved in RNA editing has been observed in mitochondria of P. polycephalum and also in kinetoplasts of Trypanosomes and could also have established two distinct activities in non-flowering plants. One of these, the U to C activity, may subsequently have been lost in the lineage of the flowering plants. 7.3. Is a CTP synthase involved? An analogous problem arises when the family of CTP synthases is considered as alternative evolutionary source for the RNA editing enzyme in plant mitochondria. CTP synthases might have been adapted to perform the amination reaction changing U to C in polynucleotides, but are unlikely to be able to catalyze the cytidine deamination. So far none of the CTP synthase enzymes described for the mononucleotide amination from UTP to CTP has been M. Takenaka et al. / Mitochondrion 8 (2008) 35–46 reported to also catalyze the deamination step. A second problem with the CTP synthases is their requirement for low molecular weight cofactors. Generally ammonia or glutamine serve as amino group donors, which, as detailed above for the transaminases, should have been quantitatively eliminated by the extensive dialysis step from the editing-competent in vitro lysates. For the in vitro assayed C to U changes, an amino group acceptor would be required, which, considering a strictly reverse reaction scenario, would also be such a smaller molecule. Alternatively any of the various tRNA modifying enzymes such as one of the nucleotide-specific deaminases or transaminases may have evolved into the RNA editing enzyme. Less likely, but still possible are DNA repair enzymes as an evolutionary source of the editing activities in plastids and mitochondria of plants. Identification of the enzyme involved in plant mitochondrial RNA editing remains the foremost problem in this field for the near future, and several approaches by genetic analysis and experimental protein identification are presently being pursued by several laboratories around the world. 7.4. Does an RNA helicase participate? The participation of an RNA helicase in plant mitochondrial RNA editing has been proposed from the observation of a strict requirement of ATP in the in vitro reactions. Particularly the finding that in plant mitochondrial in vitro assays, this ATP can be fully substituted by any of the dNTPs (and any of the NTPs) has prompted speculations about an RNA helicase (Takenaka and Brennicke, 2003). Several members of this class of enzymes have been shown to tolerate such a broad spectrum of nucleotides, which is rarely seen in other enzymes. RNA helicases participate in numerous processes of RNA maturation and might very well also be involved in RNA editing in plant mitochondria. One possible function of the RNA helicase would be to resolve inhibitory secondary structures in the template 43 mRNAs and present the RNA molecule in a form accessible to the specific trans-acting RNA editing factors. Another function of the RNA helicase could be to clear the RNA template from RNA binding proteins which adhere unspecifically to any RNA sequence and thus block access to the RNA editing proteins (Fairman et al., 2004). The main function of such general RNA binding proteins is of course the protection of the RNA against nucleases. The RNA helicase would dissociate the bound unspecific proteins along the RNA molecule, the energy which it requires for this task provided by the nucleotidetriphosphates which have to be added in vitro. This cleaning of the RNA would free the recognition site and the nucleotide to be edited and, thus, allow access of the editing activity (Fig. 5). This involvement of an RNA helicase seems to be the most likely scenario to explain the various observations made for NTP requirements in plant mitochondrial and plastid in vitro assays. In the latter the optimal concentration of the NTP/dNTP which has to be added in vitro appears to vary depending on the plant species and sometimes on the lysate preparation. The substitution of the NTP/dNTP requirement in the mitochondrial in vitro reaction by the GDH blocker phosphinothricin suggests that the observed strict nucleotide dependence of the reaction is indeed due to this auxiliary reaction of an RNA helicase, rather than a direct requirement of NTP as an energy source for the biochemical deamination reaction of C to U (Fig. 5). Participation of an RNA helicase as a separate but necessary auxiliary function in the RNA editing complex in plant mitochondria thus appears likely, but will have to be confirmed by direct identification of the RNA helicase enzyme involved. 8. Outlook on RNA editing in plant mitochondria The still open question, which has not been clarified in the 15 years since RNA editing in plant mitochondria has been first recognized, is which enzyme(s) is/are involved in the C to U alterations in the flowering plants and what Fig. 5. Cartoon of our current model of the process of RNA editing in plant mitochondria. The C to-be-edited is indicated in the RNA template molecule. The RNA is represented by a horizontal line. The trans-factor recognizes the sequence region located between 5 and 10 to about 20 and 25 nucleotides upstream of the edited C. This specificity factor cannot access the RNA which is covered by unspecific RNA binding proteins, the most prominent of which is in vitro the enzyme glutamate dehydrogenase (GDH). The GDH can be directly inhibited from binding to the RNA by the specific blocker phosphinothricin or by NADH. The analogous effect of added NTP or dNTP such as ATP or dGTP is either generated also directly through binding to the GDH or is indirectly mediated by activation of an RNA helicase. The activated helicase removes the GDH and other RNA binding proteins attached to the template RNA and makes this molecule accessible for the RNA editing activity. 44 M. Takenaka et al. / Mitochondrion 8 (2008) 35–46 is the underlying biochemistry? Is the C to U transition in flowering plants a straightforward deamination reaction? Or is a transamination process involved which moves the amino group to large acceptor molecules? Could such acceptor molecules be unspecific or dedicated RNA molecules? Are modified tRNA modifying enzymes involved such as specific deaminases which have lost their tRNA specificity? Or has one or more of the DNA repair enzymes altered its structure to also accept RNA as substrate? Connected with this unresolved problem is the question how the U to C alteration is performed in the mitochondrial (and plastid) RNAs of non-flowering plants? Recognition of the PPR protein family seems to have brought the question of the mediators of specificity in RNA editing to a realistic solution. Although these proteins are presently the most likely candidates for the specificity-mediating trans-factors, their actual involvement in RNA editing in plant mitochondria and their individual assignment to specific RNA editing sites now needs to be proven experimentally. The identity and structure of the components of the potential ‘editosome’ have to be resolved. For example, it will have to be clarified, if and how the PPR proteins interact with the enzyme? Are other proteins involved and if so, which are these? Is an RNA helicase directly involved in the RNA editing machinery? Do unspecific RNA binding proteins such as the RBP proteins, which are abundant in plant mitochondria (Vermel et al., 2002) and presumably can bind to RNA unspecifically, play a role in editing? Answers to these questions are steps towards the assembly of a model of RNA maturation in plant mitochondria. This will have to include information of the location of the mRNA molecules and the order of the various processes in the mitochondrial organelle. The model will have to assign the in vivo coupling and channeling of transcription and RNA maturation (and possibly translation) within the organelle to understand how the protein products required for plant mitochondrial functions are provided by the organelle. Acknowledgements We are grateful to the editors and the anonymous reviewer for very helpful comments on our presentation. Mizuki Takenaka was a fellow of the JSPS and Johannes Andreas van der Merwe was a fellow of the DAAD. Grants from the Deutsche Forschungsgemeinschaft supported the work in the laboratory of the authors. References Araya, A., Domec, C., Begu, D., Litvak, S., 1992. An in vitro system for the editing of ATP synthase subunit 9 mRNA using wheat mitochondrial extracts. Proc. Natl. Acad. Sci. USA 89, 1040–1044. Benne, R., Van den Burg, J., Brakenhoff, J.P.J., Sloof, P., Van Boom, J.H., Tromp, M.C., 1986. Major transcript of the frameshifted coxII gene from trypanosome mitochondria contains four nucleotides not encoded in the DNA. Cell 46, 819–826. Bentolila, S., Chateigner-Boutin, A.-L., Hanson, M.R., 2005. Ecotype allelic variation in C-to-U editing extent of a mitochondrial transcript identifies RNA-editing quantitative trait loci in Arabidopsis. Plant Physiol. 139, 2006–2016. Bergthorsson, U., Adams, K., Thomason, B., Palmer, J.D., 2003. Widespread horizontal transfer of mitochondrial genes in flowering plants. Nature 424, 197–201. Blanchard, J.L., Schmidt, G.W., 1995. Pervasive migration of organellar DNA to the nucleus in plants. J. Mol. Evol. 41, 397–406. Bock, R., Herrmann, M., Fuchs, M., 1997. Identification of critical nucleotide positions for plastid RNA editing-site recognition. RNA 3, 1194–1200. Bock, R., Herrmann, M., Kössel, H., 1996. In vivo dissection of cis-acting determinants for plastid RNA editing. EMBO J. 15, 5052–5059. Bolle, N., Kempken, F., 2006. Mono- and dicotyledonous plant-specific RNA editing sites are correctly edited in both in organello systems. FEBS Lett. 580, 4443–4448. Brennicke, A., Grohmann, L., Hiesel, R., Knoop, V., Schuster, W., 1993. The mitochondrial genome on its way to the nucleus: Different stages of gene transfer in plants. FEBS Lett. 325, 140–145. Carillo, C., Bonen, L., 1997. RNA editing status of nad7 intron domains in wheat mitochondria. Nucleic Acids Res. 25, 403–409. Carter, C.W., 1998. Nucleoside deaminases for cytidine and adenosine: comparisons with deaminases acting on RNA. In: Grosjean, H., Benne, R. (Eds.), Modification and Editing of RNA. ASM Press, Washington, DC, pp. 363–376. Chang, B.H.-J., Lau, P.P., Chan, L., 1998. Apolipoprotein B mRNA editing. In: Grosjean, H., Benne, R. (Eds.), Modification and Editing of RNA. ASM Press, Washington, DC, pp. 325–342. Chateigner-Boutin, A.-L., Hanson, M.R., 2002. Cross-competition in transgenic chloroplasts expressing single editing sites reveals shared cis elements. Mol. Cell. Biol. 22, 8448–8456. Chaudhuri, S., Maliga, P., 1996. Sequences directing C to U editing of the plastid psbL mRNA are located within a 22 nucleotide segment spanning the editing site. EMBO J. 15, 5958–5964. Chaudhuri, S., Carrer, H., Maliga, P., 1995. Site-specific factor involved in the editing of the psbL mRNA in tobacco plastid. EMBO J. 14, 25951– 25957. Choury, D., Araya, A., 2006. RNA editing site recognition in heterologous plant mitochondria. Curr. Genet. 50, 405–416. Choury, D., Farré, J.-C., Jordana, X., Araya, A., 2004. Different patterns in the recognition of editing sites in plant mitochondria. Nucleic Acids Res. 32, 6397–6406. Covello, P.S., Gray, M.W., 1989. RNA editing in plant mitochondria. Nature 341, 662–666. Covello, P.S., Gray, M.W., 1993. On the evolution of RNA editing. Trends Genet 9, 265–268. Fairman, M.E., Maroney, P.A., Wang, W., Bowers, H.A., Gollnick, P., Nilsen, T.W., Jankowsky, E., 2004. Protein displacement by DexH/ D ‘‘RNA helicases’’ without duplex unwinding. Science 304, 730– 734. Faivre-Nitschke, S.E., Grienenberger, J.M., Gualberto, J.M., 1999. A prokaryotic-type cytidine deaminase from Arabidopsis thaliana: gene expression and functional characterization. Eur. J. Biochem. 263, 896– 903. Farré, J.-C., Araya, A., 2001. Gene expression in isolated plant mitochondria: high fidelity of transcription, splicing and editing of a transgene product in electroporated organelles. Nucleic Acids Res. 29, 2484–2491. Farré, J.-C., Leon, G., Jordana, X., Araya, A., 2001. Cis recognition elements in plant mitochondrion RNA editing. Mol. Cell. Biol. 21, 6731–6737. Giegé, P., Brennicke, A., 1999. RNA editing in Arabidopsis mitochondria effects 441 C to U changes in ORFs. Proc. Natl. Acad. Sci. USA 96, 15324–15329. Gott, J.M., Visomirski-Robic, L.M., 1998. RNA editing in Physarum mitochondria. In: Grosjean, H., Benne, R. (Eds.), Modification and Editing of RNA. ASM Press, Washington, DC, pp. 395–412. M. Takenaka et al. / Mitochondrion 8 (2008) 35–46 Greeve, J., Navaratnam, N., Scott, J., 1991. Characterization of the apolipoprotein B mRNA editing enzyme: no similarity to the proposed mechanism of RNA editing in kinetoplastid protozoa. Nucleic Acids Res. 19, 3569–3576. Grohmann, L., Thieck, O., Herz, U., Schröder, W., Brennicke, A., 1994. Translation of nad9 mRNAs in mitochondria from Solanum tuberosum is restricted to completely edited transcripts. Nucleic Acids Res. 22, 3304–3311. Groth-Malonek, M., Knoop, V., 2005. Bryophytes and other basal land plants: the mitochondrial perspective. Taxon 54, 293–297. Gualberto, J.M., Lamattina, L., Bonnard, G., Weil, J.-H., Grienenberger, J.M., 1989. RNA editing in wheat mitochondria results in the conservation of protein sequences. Nature 341, 660–662. Hajduk, S.L., Sabatini, R.S., 1998. Mitochondrial mRNA editing in kinetoplastid protozoa. In: Grosjean, H., Benne, R. (Eds.), Modification and Editing of RNA. ASM Press, Washington, DC, pp. 377–394. Handa, H., 2003. The complete nucleotide sequence and RNA editing content of the mitochondrial genome of rapeseed (Brassica napus L.): comparative analysis of the mitochondrial genomes of rapeseed and Arabidopsis thaliana. Nucleic Acids Res. 31, 5907–5916. Hegemann, C.E., Hayes, M.L., Hanson, M.R., 2005. Substrate and cofactor requirements for RNA editing of chloroplast transcripts in Arabidopsis in vitro. Plant J. 42, 124–132. Hernould, M., Suharsono, S., Litvak, S., Araya, A., Mouras, A., 1993. Male-sterility induction in transgenic tobacco plants with an unedited atp9 mitochondrial gene from wheat. Proc. Natl. Acad. Sci. USA 90, 2370–2374. Hiesel, R., Wissinger, B., Schuster, W., Brennicke, A., 1989. RNA editing in plant mitochondria. Science 246, 1632–1634. Hirose, T., Sugiura, M., 2001. Involvement of a site-specific trans-acting factor and a common RNA-binding protein in the editing of chloroplast RNA: development of an in vitro RNA editing system. EMBO J. 20, 1144–1152. Knoop, V., 2004. The mitochondrial DNA of plants: peculiarities in phylogenetic perspective. Curr. Genet. 46, 123–139. Kotera, E., Tasaka, M., Shikanai, T., 2005. A pentatricopeptide repeat protein is essential for RNA editing in chloroplasts. Nature 433, 326– 330. Kubo, N., Kadowaki, K., 1997. Involvement of 5’ flanking sequence for specifying RNA editing sites in plant mitochondria. FEBS Lett. 413, 40–44. Lippok, B., Wissinger, B., Brennicke, A., 1994. Differential RNA editing in closely related introns in Oenothera mitochondria. Mol. Gen. Genet. 243, 39–46. Lurin, C., Andrés, C., Aubourg, S., Bellaoui, M., Bitton, F., Bruyère, C., Caboche, M., Debast, C., Gualberto, J.M., Hoffmann, B., Lecharny, A., Le Ret, M., Martin-Magniette, M.-L., Mireau, H., Peeters, N., Renou, J.-P., Szurek, B., Taconnat, L., Small, I., 2004. Genome-wide analysis of Arabidopsis pentatricopeptide repeat proteins reveals their essential role in organelle biogenesis. Plant Cell 16, 2089–2103. Mahendran, R., Spottswood, M.R., Miller, D.L., 1991. RNA editing by cytidine insertion in mitochondria of Physarum polycephalum. Nature 349, 434–438. Malek, O., Lättig, K., Hiesel, R., Brennicke, A., Knoop, V., 1996. RNA editing in bryophytes and a molecular phylogeny of land plants. EMBO J. 15, 1403–1411. Marchfelder, A., Binder, S., 2004. Plastid and plant mitochondrial RNA processing and RNA stability. In: Daniell, H., Chase, C.D. (Eds.), Molecular Biology and Biotechnology of Plant Organelles. Springer, Dordrecht, pp. 261–294. Miyamoto, T., Obokata, J., Sugiura, M., 2002. Recognition of RNA editing sites is directed by unique proteins in chloroplasts: biochemical identification of cis-acting elements and trans-acting factors involved in RNA editing in tobacco and pea chloroplasts. Mol. Cell. Biol. 22, 6726–6734. Miyamoto, T., Obokata, J., Sugiura, M., 2003. A site-specific factor interacts directly with its cognate RNA editing site in chloroplast transcripts. Proc. Natl. Acad. Sci. USA 101, 48–52. 45 Mulligan, M., 2004. RNA editing in plant organelles. In: Daniell, H., Chase, C.D. (Eds.), Molecular Biology and Biotechnology of Plant Organelles. Springer, Dordrecht, pp. 239–260. Navaratnam, N., Fujino, T., Bayliss, J., Jarmuz, A., How, A., Richardson, N., Somasekaram, A., Bhattacharya, S., Carter, C., Scott, J., 1998. Escherichia coli cytidine deaminase provides a molecular model for ApoB RNA editing and a mechanism for RNA substrate recognition. J. Mol. Biol. 275, 695–714. Neuwirt, J., Takenaka, M., van der Merwe, J.A., Brennicke, A., 2005. An in vitro RNA editing system from cauliflower mitochondria: editing site recognition can vary between different plant species. RNA 11, 1563–1570. Notsu, Y., Masood, S., Nishikawa, T., Kubo, N., Akiduki, G., Nakazono, M., Hirai, A., Kadowaki, K., 2002. The complete sequence of the rice (Oryza sativa L.) mitochondrial genome: frequent DNA acquisition and loss during the evolution of flowering plants. Mol. Genet. Genomics 268, 434–445. O’Connell, M.A., 1997. RNA editing: rewriting receptors. Curr. Biol. 7, R437–R439. Oda, K., Yamato, K., Ohta, E., Nakamura, Y., Takemura, M., Nozato, N., Akashi, K., Kanegae, T., Ogura, Y., Kohchi, T., Ohyama, K., 1992. Gene organization deduced from the complete sequence of liverwort Marchantia polymorpha mitochondrial DNA: a primitive form of plant mitochondrial genome. J. Mol. Biol. 223, 1–7. Okuda, K., Myouga, F., Motohashi, R., Shinozaki, K., Shikanai, T., 2007. Conserved domain structure of pentatricopeptide repeat proteins involved in chloroplast RNA editing. Proc. Natl. Acad. Sci. USA 104, 8178–8183. Phreaner, C.G., Williams, M.A., Mulligan, R.M., 1996. Incomplete editing of rps12 transcripts results in the synthesis of polymorphic polypeptides in plant mitochondria. Plant Cell 8, 107–117. Pring, D.R., Mullen, J.A., Kempken, F., 1992. Conserved sequence blocks 5’ to start codons of plant mitochondrial genes. Plant Mol. Biol. 19, 313–317. Rajasekhar, V.K., Mulligan, R.M., 1993. RNA editing in plant mitochondria: a-phosphate is retained during C-to-U conversion in mRNAs. Plant Cell 5, 1843–1852. Reed, M.L., Peeters, N.M., Hanson, M.R., 2001. A single alteration 20 nt 5 0 to an editing target inhibits chloroplast RNA editing in vivo. Nucleic Acids Res. 29, 1507–1513. Schuster, W., Brennicke, A., 1989. Conserved sequence elements at putative processing sites in plant mitochondria. Curr. Genet. 14, 187– 192. Shinakai, T., 2006. RNA editing in plant organelles: machinery, physiological function and evolution. Cell. Mol. Life Sci. 63, 698–708. Small, I., Peeters, N., 2000. The PPR motif—a TPR-related motif prevalent in plant organellar proteins. Trends Biochem. Sci. 25, 46–47. Sommer, B., Köhler, M., Sprengel, R., Seeburg, P., 1991. RNA editing in brain controls a determinant of ion flow in glutamate-gated channels. Cell 67, 11–19. Späth, B., Settele, F., Schilling, O., D’Angelo, I., Vogel, A., Feldmann, I., Meyer-Klaucke, W., Marchfelder, A. 2007. Metal requirements and phosphodiesterase activity of tRNase Z enzymes. Biochemistry, in press. Staudinger, M., Kempken, F., 2003. Electroporation of isolated higherplant mitochondria: transcripts of an introduced cox2 gene, but not an atp6 gene, are edited in organello. Mol. Genet. Genomics 269, 553–561. Staudinger, M., Bolle, N., Kempken, F., 2005. Mitochondrial electroporation and in organello RNA editing of chimeric atp6 transcripts. Mol. Genet. Genomics 273, 130–136. Stupar, R.M., Lilly, J.W., Town, C.D., Cheng, Z., Kaul, S., Buell, C.R., Jiang, J., 2001. Complex mtDNA constitutes an approximate 620-kb insertion on Arabidopsis thaliana chromosome 2: implication of potential sequencing errors caused by large-unit repeats. Proc. Natl. Acad. Sci. USA 98, 5013–5099. Takenaka, M., Brennicke, A., 2003. In vitro RNA editing in pea mitochondria requires NTP or dNTP, suggesting involvement of an RNA helicase. J. Biol. Chem. 278, 47526–47533. 46 M. Takenaka et al. / Mitochondrion 8 (2008) 35–46 Takenaka, M., Neuwirt, J., Brennicke, A., 2004. Complex cis-elements determine an RNA editing site in pea mitochondria. Nucleic Acids Res. 32, 4137–4144. Takenaka, M., Verbitskiy, D., van der Merwe, J.A., Zehrmann, A., Plessmann, U., Urlaub, H., Brennicke, A., 2007. In vitro RNA editing in plant mitochondria does not require added energy. FEBS Lett. 581, 2743–2747. Turmel, M., Otis, C., Lemieux, C., 2003. The mitochondrial genome of Chara vulgaris: insights into the mitochondrial DNA architecture of the last common ancestor of green algae and land plants. Plant Cell 15, 1888–1903. van der Merwe, J.A., Takenaka, M., Neuwirt, J., Verbitskiy, D., Brennicke, A., 2006. RNA editing sites in plant mitochondria can share cis-elements. FEBS Lett. 580, 268–272. Verbitskiy, D., Takenaka, M., Neuwirt, J., van der Merwe, J.A., Brennicke, A., 2006. Partially edited RNAs are intermediates of RNA editing in plant mitochondria. Plant J. 47, 408–416. Vermel, M., Guermann, B., Delage, L., Grienenberger, J.M., MarechalDrouard, L., Gualberto, J.M., 2002. A family of RRM-type RNAbinding proteins specific to plant mitochondria. Proc. Natl. Acad. Sci. USA 99, 5866–5871. Yu, W., Schuster, W., 1995. Evidence for a site-specific cytidine deamination reaction involved in C to U editing in plant mitochondria. J. Biol. Chem. 270, 18227–18233. Zabaleta, E., Mouras, A., Hernould, M., Suharsono, S., Araya, A., 1996. Transgenic male-sterile plant induced by an unedited atp9 gene is restored to fertility by inhibiting its expression with antisense RNA. Proc. Natl. Acad. Sci. USA 93, 11259–11263.