This Article Abstract Full Text (PDF) All Versions of this Article: 139/4/2006    most recent pp.105.069013v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Web of Science (12) Citing Articles via Google Scholar Google Scholar Articles by Bentolila, S. Articles by Hanson, M. R. Search for Related Content PubMed PubMed Citation Articles by Bentolila, S. Articles by Hanson, M. R. Agricola Articles by Bentolila, S. Articles by Hanson, M. R.

GENETICS, GENOMICS, AND MOLECULAR EVOLUTION

Ecotype Allelic Variation in C-to-U Editing Extent of a Mitochondrial Transcript Identifies RNA-Editing Quantitative Trait Loci in Arabidopsis¹

Department of Molecular Biology and Genetics, Cornell University, Ithaca, New York 14853

	ABSTRACT

TOP ABSTRACT RESULTS DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

 
In higher plants, RNA editing is a posttranscriptional process that converts C to U in organelle mRNAs. Although RNA editing in mitochondria occurs much more frequently than in chloroplasts, editing of exogenously supplied RNA substrates in vitro and in organello has shown that editing in the two organelles shares some common features. In particular, the 20 nucleotides upstream of the editing site play an important role in specifying the C to be edited. Biochemical approaches have allowed the identification of features of cis-sequences necessary for RNA editing to occur, but have failed to identify any of the components of the mitochondrial editing machinery. In order to implement a genetic approach for identification of editing factors, we have identified a polymorphism in the editing efficiency of a mitochondrial site between two ecotypes of Arabidopsis (Arabidopsis thaliana), Columbia (Col) and Landsberg erecta (Ler). In rosette leaves, an editing site within the ccb206 mitochondrial gene is more highly edited in Col than in Ler. Depending on the development stage and tissue analyzed, the difference in editing extent varies between the two ecotypes; for example, in floral buds, editing extent does not differ. Single-point regression analysis of the editing efficiency in a sample of recombinant inbred lines derived from a cross between Col and Ler allowed the identification of two quantitative trait loci controlling this trait. A member of the pentatricopeptide repeat protein family that carries a putative mitochondrial transit sequence has been identified near a major quantitative trait locus on chromosome 4.

Progress has recently been made in the characterization of chloroplast editing cis- and trans-factors since both in vivo (Chaudhuri and Maliga, 1996) and in vitro systems (Hirose and Sugiura, 2001; Miyamoto et al., 2002; Hegeman et al., 2005) have become available. Sequences important in the selection of the editable C have been found 5' of C targets of editing in chloroplasts. Furthermore, clusters of two to five editing sites that share similar 5' sequences have been identified. Evidence for functional significance of these clusters comes from their coinhibition in editing efficiency as a consequence of overexpression of one cluster member (Chateigner-Boutin and Hanson, 2002) as well as their developmental coregulation (Chateigner-Boutin and Hanson, 2003). A current hypothesis is that the editing sites within the same cluster share a nuclear-encoded protein recognition factor that may be expressed in limited quantities in some tissues at particular times in development, thus affecting editing efficiency.

In mitochondria, the lack of transformation methods and the absence of a reliable in vitro or in organello system, until recently, as well as the large number of editing sites, have slowed progress in identifying components of the molecular apparatus responsible for RNA editing. Electroporation of isolated wheat mitochondria with RNA substrates has revealed that, in mitochondria as in chloroplasts, 5' sequences are important for specification of editing (Farré et al., 2001; Choury et al., 2004). The cis-requirements for the editing of a site in atp9 mRNA have also been determined by editing within a pea mitochondrial extract (Takenaka and Brennicke, 2003; Takenaka et al., 2004). Results from these studies confirm the prominent role of the 20 nucleotides upstream of the editing site. Mitochondrion and plastid editing systems share some other similarities. The 5' portion of several mitochondrial site sequences exhibit similarities to putative cis-elements found in chloroplasts (Chateigner-Boutin and Hanson, 2002). This finding raised the possibility of shared recognition factors encoded by the nucleus and dually targeted to both organelles. However, it is unlikely that all factors are shared between the organelles, as it has been found that the editing sites within a portion of the cox2 mitochondrial gene transcribed in transgenic chloroplasts were not edited by the chloroplast machinery (Sutton et al., 1995). Furthermore, presumably many more editing factors are needed in mitochondria than in chloroplasts.

Recently, an Arabidopsis gene that affects the editing efficiency of a chloroplast editing site has been identified (Kotera et al., 2005). While studying mutants affected in NADH dehydrogenase activity, the authors found that one such mutated allele, ccr4, also reduced editing of ndhD. The authors performed map-based cloning to discover that CRR4 is a member of the pentatricopeptide repeat (PPR) gene family (Small and Peeters, 2000). CRR4 is closely related in structure to CRR2, another PPR-containing gene involved in the intergenic RNA cleavage between rps7 and ndhB (Hashimoto et al., 2003). The only structural difference between the proteins encoded by these two genes resides in a C-terminal 15-amino acid motif that is less conserved in CRR2. This 15-amino acid motif is too small to carry the catalytic activity of cytidine deamination that is necessary for the RNA-editing process. The authors therefore propose that CRR4 is a site recognition factor that interacts with the target RNA to recruit an editing enzyme, analogous to the interaction of the RNA-binding protein apobec-1 complementation factor (ACF) with a cytidine deaminase, APOBEC1, to the mammalian apoB editing site (Keegan et al., 2001).

A possible approach to identify editing factors is the genetic mapping of an editing polymorphism between two genotypes, followed by map-based cloning of the gene responsible for the polymorphism. Here, we present the identification of an editing polymorphism between two ecotypes of Arabidopsis, Columbia (Col) and Landsberg erecta (Ler). The polymorphism occurs at a particular site within the mitochondrial transcript ccb206 (orf206), which encodes a protein similar to a component of a heme transporter involved in cytochrome c biogenesis in photosynthetic bacteria (Schuster, 1994). This site is differentially edited between the two ecotypes, depending on the developmental stage and particular tissue. Recombinant inbred lines (RILs) generated from a cross between these two ecotypes allowed us to locate two quantitative trait loci (QTLs) linked to the editing efficiency of this site.

	RESULTS

TOP ABSTRACT RESULTS DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

 

Editing Sites within ccb206 Transcripts in Mitochondria

In order to find an editing site exhibiting a differential editing efficiency between the two ecotypes of Arabidopsis, Col and Ler, we examined a gene that exhibits numerous editing sites. According to Giege and Brennicke (1999), the gene ccb206 (NP_085482) has the largest number of edited Cs in Arabidopsis mitochondria; its open reading frame of 621 nucleotides was reported to contain 39 sites.

We examined the editing extent of all sites within ccb206 by bulk sequencing of reverse transcription (RT)-PCR products obtained from rosette leaves and compared the cDNAs to the genomic sequence. Three sites originally described in Giege and Brennicke's (1999) survey of ccb206 editing, which was performed on cultured cells of Arabidopsis ecotype Col, were not edited in rosette leaves. Furthermore, we found that ccb206 transcripts exhibited an editing site not previously reported in Arabidopsis but described in rapeseed (Itani and Handa, 1998; Tables I and II). In Col and Ler rosette leaves, 37 Cs are modified to Us in ccb206 transcripts.

In the raw sequence data, the site ccb206 C24 attracted our attention; at this position, the C/T was equally represented in Col, but C was prominent in Ler. ccb206 C24 is located in the first position within the codon CUG, modified to UUG by editing. Because both codons encode a Leu residue (position 136), C24 is a silent site. We confirmed that this site is partially edited in rosette leaves of both ecotypes by poisoned primer extension (PPE; Fig. 1). We found that C24 is edited in Ler at an average of 23% and in Col at 51%. This site is differentially edited in the two ecotypes despite identical genomic sequences of the ccb206 gene and comparable levels of the ccb206 transcript (data not shown). This is in contrast to a polymorphism previously detected in editing of a site in petunia (Petunia hybrida) nad3. nad3 transcripts were much less abundant in a genotype exhibiting less editing of a particular C than in a genotype where the editing extent was higher (Lu and Hanson, 1992). In the petunia example, nad3 transcripts in plants exhibiting the less edited, less abundant phenotype may have been less stable, so that turnover occurred before editing could be completed.

View larger version (35K): [in this window] [in a new window]   Figure 1. Editing extent of ccb206 C24 in different tissues of Arabidopsis ecotype Col and Ler. Editing efficiency was assayed by PPE. A, PPE was performed on an RT-PCR product from the radiolabeled oligo PPE-ccb206-C24; extension of the primer was poisoned by ddCTP incorporation. PPE products were resolved on a 12% sequencing gel, which was then exposed on a phosphorimager screen. B, The percentage of edited transcript was determined by quantifying the radioactivity associated with edited and unedited ccb206 C24 using ImageQuant software (Molecular Dynamics). The editing efficiency is reported on the y axis and is the average of three independent PPE reactions. The error bars represent the SDs of the three measurements. 0, PPE without template indicating the size of the radiolabeled primer; U, PPE with a cloned (genomic) unedited PCR product; E, PPE with a cloned edited RT-PCR product.

We investigated the editing efficiency of this mitochondrial site in different tissues of both ecotypes, knowing that the editing extent of partially edited sites can vary between tissues in plastids (Peeters and Hanson, 2002). We found that, in Ler, C24 was edited at 23% to 24% in all investigated tissues (Fig. 1). On the contrary, Col C24 editing efficiency varied depending on the tissue analyzed. The highest editing level was found in rosette leaves, where the site is edited at 51%, and the lowest in floral buds, where C24 is edited only at 23%. In floral buds, C24 is edited to the same extent in the two ecotypes. We decided to continue our study on rosette leaves, where the difference was the most prominent.

Dominance of the Ler Phenotype

To further analyze the genetic basis of this polymorphism, we crossed the two ecotypes in both combinations: Col or Ler as a female. In both cases, we obtained several F₁ hybrids. Their hybrid status was checked using cleaved amplified polymorphic sequence markers (Konieczny and Ausubel, 1993; data not shown). The editing extent of ccb206 C24 was investigated in rosette leaves of all F₁ hybrids. Hybrids generated from both cross combinations exhibit a Ler phenotype (Fig. 2), which is therefore dominant over the Col phenotype. These data also suggest that any genes controlling the phenotype are located in the nucleus, since no maternal effect was observed in the editing extent of C24 in F₁ hybrids.

View larger version (36K): [in this window] [in a new window]   Figure 2. Editing extent of ccb206 C24 in rosette leaves of F1s generated from a cross between Arabidopsis ecotype Col and Ler. A, PPE results obtained for the parents Col and Ler and for four F1 progeny. B, Editing efficiency of Col and Ler and of F1s generated from a cross between the two ecotypes in both combinations. The value for Col and Ler is the same as the one reported in Figure 1B and comes from three different measurements. For the hybrids, the value is the average of assays of four different F1 progeny for Col x Ler and of three different F1 progeny for Ler x Col. The error bars represent the SDS of the measurements for each genotype. 0, PPE without template indicating the size of the radiolabeled primer; U, PPE with a cloned (genomic) unedited PCR product; E, PPE with a cloned edited RT-PCR product; Col x Ler, F1 generated from a cross with Col as a female; Ler x Col, F1 generated from a cross with Ler as a female.

To map loci involved in the editing polymorphism between Col and Ler, we assayed the editing extent of a population of RILs generated from a cross between the two ecotypes, with Ler as the female parent (Lister and Dean, 1993). We analyzed the editing efficiency of ccb206 C24 in the 30 RILs selected as having the highest frequency of recombination over the five chromosomes.

The PPE experiments were repeated three times and the average editing efficiencies are represented in Figure 3. The variance within the RILs was estimated to be 2.48, while the variance between the RILs was estimated to be 77.09. A likelihood ratio statistic test showed that the variance between the RILs was highly significantly different from zero. N1974 is the only outlier found in the RIL population with an average editing percentage of 17% compared to the average editing percentage of 23% for Ler, the low-edited parent (Fig. 3). Finding only one outlier in the RIL population suggests that no QTL with an opposite effect to the ones found in the parent lines (namely, reduced editing for Ler versus increased editing for Col) was likely to be found. Some of the RILs exhibited a Col phenotype with an editing extent of C24 of about 50%, such as N1929 and N1901. Others have a Ler phenotype, such as N1966, but most of the analyzed RILs exhibit an intermediate phenotype (Fig. 3). The fact that several RILs with editing efficiency comparable to the parental lines were found in a relatively small subsample of the RIL population argues for a relatively small number of QTLs controlling the editing efficiency of ccb206 C24.

View larger version (35K): [in this window] [in a new window]   Figure 3. Editing extent of ccb206 C24 in 30 RILs generated from a cross between Col and Ler. The average of the editing percentage obtained from three different PPE experiments was determined as indicated in Figure 1. The error bars represent the SDs of the three measurements.

The average editing efficiency of the RILs was used as a quantitative trait. In order to map QTLs with these RILs, we used 62 previously mapped markers. The markers span most of the genome at intervals <15 cM, except for a part of chromosome 3 (20.2 cM g4711-g4564-b) and chromosome 5 (27.2 cM g4715-b-m247; 18.3 cM g4028-m435). Identification of QTLs and estimation of their effects were obtained with three different analytical methods, single-point regression analysis (SPA), interval mapping (IM), and composite interval mapping (CIM). The three different approaches identified a major QTL on chromosome 4 colocalizing with the marker g6837 (Table III). The likelihood odds ratio (LOD) scores, which indicate how much more probable the data are to have arisen assuming the presence of a QTL than assuming its absence, are well above the empirical thresholds corresponding to an overall false positive rate of 1%. The amount of phenotypic variation explained by this QTL (R²) ranges from 38% to 48%, depending on the method used to map it (Table III). At this peak marker, the average editing of C24 in Col-Col genotypes is 39%, whereas it is only 27% for Ler-Ler genotypes. Therefore, the Col allele is associated with an increase in C24 editing efficiency of about 6%, which represents the additive effect of this QTL. With the segregating population used in this study, it is not possible to estimate the dominance effect for any QTL, since all the RILs are homozygous for their whole genome.

Multiple regression analysis with the markers g6837 and m532 as factors showed that the proportion of phenotypic variation explained by the model amounts to 61%, which corresponds to the sum of the individual R² for each marker. From this result, we can conclude that the two QTLs are likely to have an independent action on the editing of ccb206 C24.

Search for Mitochondrial Editing Site Homologs to ccb206 C24 Controlled by the Same QTLs

Because editing of ccb206 C24 is silent—editing does not change the encoded amino acid sequence—this site is likely to have no physiological importance. Why should silent editing occur? One possibility is that a recognition factor that operates on a nonsilent site happens to interact with C24 because of a fortuitous similarity in recognition cis-sequences between the two sites. The editing of C24 would then be a by-product resulting in editing that fails to alter the protein sequence. ccb206 C24 could possibly cluster with more physiologically important sites, which would need editing to change the encoded amino acid residues. We attempted to find editing sites with a 5' sequence similar to the one of ccb206 C24 by using the 20 nucleotides upstream of the C target of editing in a search for short, nearly identical, sequences in the complete Arabidopsis mitochondrial editing database. Figure 4 shows a subset of the sites identified by Pratt software that we decided to investigate further. All these editing sites are nonsilent.

View larger version (48K): [in this window] [in a new window]   Figure 4. Editing sites sharing some similarity in their putative cis-recognition sequences with ccb206 C24. These sites were identified by using Pratt software (see "Materials and Methods" for details). ccb206 C24 is shown in the top line for reference. Identical nucleotides shared by other sites are shown in gray-shaded background. Cs targeted for editing are in bold. The number in parenthesis refers to the nucleotide position of the edited C in the coding sequence of each gene A from the initiation codon having the position 1. *, orfx and ccb203 coding sequences are incomplete and do not start with an ATG codon.

Differential Processing of the ccb206 Transcript

Although we did not find a difference in the accumulation of the ccb206 transcript in Col and Ler, a different RNA-banding pattern was observed between the two ecotypes upon northern analysis (Fig. 5). It is thus possible that processing of the ccb206 transcript specific to each ecotype is causing the observed difference in ccb206 C24 editing. This hypothesis was tested by assessing the ccb206 transcript pattern in eight RILs, four highly edited as Col and four poorly edited as Ler (Fig. 5). At least three of these RILs show a ccb206 RNA pattern contradicting this hypothesis. N1946, highly edited as Col, shows the same pattern as Ler. N1974 and N1984, poorly edited as Ler, show the same pattern as Col. (The pattern of N1929 is difficult to assess unequivocally on the northern presented in Fig. 5.) We can conclude from these data that processing of the ccb206 transcript and editing of the ccb206 C24 are independent.

View larger version (65K): [in this window] [in a new window]   Figure 5. The ccb206 transcript pattern does not cosegregate with ccb206 C24 editing efficiency. Numbers below the figure refer to the editing extent of ccb206 C24.

	DISCUSSION

TOP ABSTRACT RESULTS DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

Editing of this particular site, ccb206 C24, is partial in both ecotypes; however, in Col, but not in Ler, the editing extent is differentially regulated in the tissues and developmental stages investigated. Crosses between the two ecotypes revealed that the Ler phenotype is dominant over the Col phenotype and that the genes involved in this polymorphism are located in the nucleus.

One puzzling feature of the polymorphism we identified is that the less edited phenotype is dominant to the more edited phenotype. If the polymorphism reflects a difference in efficiency of recognition of a site or the ability to attract an editing enzyme to a site, then possibly the higher editing phenotype would be dominant. However, editing factors are present in limited quantities, as shown by the reduction in editing of endogenous genes when a plastid transgene carrying an editing site is overexpressed (Chateigner-Boutin and Hanson, 2002). Furthermore, members of the PPR gene family, some of which may encode editing factors, are known to exhibit transcripts with very low abundance (Lurin et al., 2004). It is possible that the dominant allele encodes a trans-factor that is more abundant than the factor encoded by the recessive allele. Furthermore, the presence of a trans-factor encoded by the dominant allele might result in low editing because it occupies the editing substrate, but does not bind as well to an editing enzyme that is required for completion of the editing reaction. According to this hypothesis, one or both QTL could represent factors that recognize editing sites despite the dominance of the less edited phenotype.

Alternatively, the QTLs could represent inhibitory factors that are directly or indirectly affecting editing efficiency. There is precedence for an inhibitory factor affecting C-to-U editing in the apoB gene, which is affected by the cytidine deaminase APOBEC-1 and ACF, an accessory factor. A regulatory factor termed GRY-RBP is an RNA-binding protein that shares 50% identity with ACF and inhibits both the binding of ACF and the editing event (Blanc et al., 2001). In contrast to such a direct effect on editing, possibly the QTL might influence RNA-editing efficiency indirectly by modulating the abundance of a mitochondrial RNA species that competes for the ccb206 C24 RNA. Perhaps a transcript carrying an editing site that clusters with the ccb206 C24 sequence is more highly expressed in Ler than in Col, leading to reduced editing because of a limited amount of the cluster's editing factors. The competing transcript's abundance could be regulated by one or both of the QTLs. A number of nuclear genes have been identified that control plant organelle transcript and protein abundance (for review, see Stern et al., 2004; Hanson and Bentolila, 2004).

We searched for a nonsilent editing site that could be the primary target of the QTLs identified in this study. Because ccb206 C24 is a silent editing site that does not change the amino acid sequence encoded by the DNA, we speculated that there might be another nonsilent editing site sharing some sequence similarity in its upstream sequence with ccb206 C24. Only one of the tested sites, ccb203 (320), showed a significant difference between the Col and the Ler ecotypes (Table IV). The editing assessment of some of the RILs for ccb203 (320) did not support this polymorphism being controlled by the same QTLs as ccb206 C24. Several reasons might explain why we have failed to identify a site that might be affected by the same QTLs as ccb206 C24. We investigated only the most promising sites, based on visual inspection, among the sites identified by Pratt software. For instance, orfx (409), which has the closest match to ccb206 C24, shares with the latter 13 identical nucleotides in its 20 upstream nucleotides (Fig. 4). Editing of the C in position –2 of orfx (409) makes its upstream sequence even more similar to ccb206 C24. The same observation can be made for the sites ccb452 (155) and (415), where editing of the C at position –9 makes the upstream sequences of these sites more similar to ccb206 C24. Inversely, editing can decrease the similarity of an upstream sequence with the one from ccb206 C24 as seen for rps4 (1,057; edited Cs at positions –5 and –15). Some upstream sequences of the sites we investigated may appear to be poorly related to ccb206 C24 because we have presented them without gaps for the purpose of simplicity (Fig. 4). But gaps were allowed in our screen with Pratt software. For instance, the introduction of a gap in the upstream sequence of ccb206 C24 between the T at position –4 and the C at position –5 restores a contiguous stretch of six identical nucleotides (CTTGAC) between ccb206 C24 and nad4 (896). Gaps occur within the alignments of the upstream regions of known chloroplast editing-site clusters (Chateigner-Boutin and Hanson, 2002, 2003). The possible existence of gaps makes it difficult to select all possible candidate sites that might be related to ccb206 C24.

Another explanation for the failure to find another editing site functionally related to ccb206 C24 is our choice to screen the published editing database comprising 456 editing sites (Giege and Brennicke, 1999). As acknowledged by the authors, this database is not complete. About 31 kb of open reading frames >100 amino acids and 17 kb of introns have not been analyzed. Moreover, mitochondrial RNA was extracted from a cell-suspension culture in the study by Giege and Brennicke (1999). Editing is dependent upon the tissue analyzed and the environmental conditions in chloroplasts (Karcher and Bock, 1998; Nakajima and Mulligan, 2001; Peeters and Hanson, 2002) and in mitochondria (this study; Kurihara-Yonemoto and Handa, 2001). Some sites might thus have been overlooked because they are not edited in suspension cells. Although it is difficult to evaluate the number of such sites, it might be rather high. In an independent study covering 150 reported sites in seven mitochondrial genes, we found 17 new, fully edited sites and 22 new, partially edited sites (S. Bentolila, unpublished data).

A BLAST search for nearly identical short sequences to the 20 nucleotides upstream of the ccb206 C24 in the Arabidopsis mitochondrial genome did not reveal any additional promising sequences to investigate further (data not shown). As discussed earlier, because editing can itself alter the RNA substrate that is targeted for further editing, possibly an RNA sequence highly similar to ccb206 C24 could exist but not be recognizable by searching for matches to unedited genomic DNA.

Having failed to uncover a functional nonsilent site related to ccb206 C24 does not invalidate our hypothesis about the existence of such a site. One possible reason we have not detected a polymorphism in a nonsilent site is that both of the proteins encoded by the QTL recognize the nonsilent site with strong affinity. If so, then high editing of the nonsilent site would occur in both ecotypes. If an accidental recognition of ccb206 C24 is occurring, then the lower affinity of both factors for the silent site might have exposed differential binding of two factors encoded by different alleles, resulting in a differential editing. Thus, one of the sites identified by Pratt software and shown in Table IV might actually be the real target of the major QTL.

Could there be a selective advantage to editing a silent site? One possible scenario relates to the codon usage in the Arabidopsis mitochondrion, which shows a nearly systematic preference for codons ending with U for a specific amino acid (Giege and Brennicke, 1999). It has been speculated that this bias could partly explain the persistence of silent editing affecting the third position. However, while this theory might be valid as a factor in the persistence of silent editing at the species level, it is probably not relevant to the polymorphism we observed between the Col and Ler ecotypes. It is unlikely that the overall codon usage in the mitochondrial genome of Ler differs from codon usage in Col.

Recently, CRR4, a gene essential for the editing of a site that creates an initiation codon in the chloroplast gene ndhD, has been cloned (Kotera et al., 2005). CRR4 is a member of the PPR family, one of the largest plant gene families first described in Arabidopsis (Small and Peeters, 2000). Although PPR-containing genes are found in all eukaryotes analyzed, they have greatly expanded in plants. According to the nature of the motifs found in the PPRs, this family can be divided into two subfamilies, the P subfamily and the PLS subfamily, which is specific to plants (Lurin et al., 2004). The PLS subfamily can be further divided into four subgroups, PLS, E, E+, and DYW, depending on the C-terminal domain found after the PPR motifs. Several lines of evidence point to an involvement of the PPR proteins in the control of gene expression in organelles. Most of the PPR proteins are predicted to be targeted to either mitochondria or plastids (Lurin et al., 2004). Positional cloning of cytoplasmic male sterility restorer genes, Rf in petunia (Bentolila et al., 2002), Rfk and Rfo in radish (Brown et al., 2003; Desloire et al., 2003; Koizuka et al., 2003), and Rf-1 in rice (Komori et al., 2004) has shown that they all encode PPR proteins. Lurin et al. (2004) speculated that the PLS subfamily of PPR proteins, specific to plants, might be involved in RNA editing because this posttranscriptional process is unique to land plants. Two hundred PPR proteins belonging to the PLS subfamily have been identified in the Arabidopsis genome (Lurin et al., 2004). It is thus tempting to postulate that the QTLs reported in this study might indeed correspond to one of these PLS proteins. We looked for the presence of such a protein in the vicinity of g6837 on chromosome 4, the marker that colocalizes with the major QTL. We found one or two candidate genes for the major QTL depending on the length of the support interval obtained with the CIM, the method that gave the sharpest LOD peak of the three QTL mapping methods used in this study. The one LOD support interval for CIM, CIM1, covers 301 kb extending from the Arabidopsis Genome Initiative (AGI) map position 7,754 kb to the AGI map position 8,055 kb. In this interval, 67 predicted genes (http://www.ncbi.nlm.nih.gov/mapview/maps.cgi?TAXID=3702&MAPS=default&CHR=IV) are found. Only one PPR-containing gene, At4g13650, is found in CIM1. At4g13650, the PPR protein encoded by this gene, is predicted to be targeted to mitochondria by Predotar (Small et al., 2004) and TargetP (Emanuelsson et al., 2000), two prediction programs designed to determine the subcellular localization of proteins. The two LOD support interval for CIM, CIM2, covers 553 kb extending from the AGI map position 7,554 kb to the AGI map position 8,107 kb. CIM2 comprises 141 predicted genes, of which two are PPR-containing genes. In addition to At4g13650, CIM2 also includes At4g14050. At4g14050, like At4g13650, is predicted to be targeted to mitochondria by both prediction programs. Both of these PPR proteins belong to the DYW subclass.

Two arguments support these PPR-containing genes, At4g13650 and At4g14050, as promising candidates for the major QTL. First, their localization is predicted to be in the mitochondrion, as expected of a factor controlling the editing of a mitochondrial site. Second, both genes belong to the DYW subclass, which has recently been observed to exhibit a phylogenetic correlation with RNA editing in plants. The DYW domain was found in the Jungermanniales, the largest order of liverworts, where RNA editing is present. By contrast, the DYW domain was not found in the Marchantiales, where editing is absent (I. Small, personal communication).

	CONCLUSION

TOP ABSTRACT RESULTS DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

 
In this initial study, we have mapped two QTLs linked to the difference in editing efficiency between two ecotypes of Arabidopsis. One of these QTLs lies on chromosome 4 and plays a major role, while the other QTL lies on chromosome 1 and shows a smaller effect. Our results prove that it is possible to map major editing QTLs even with a small population. Given the level of precision attached to QTL mapping, we must undertake finer mapping in order to clone the actual genes involved in this polymorphism. We intend also to utilize a candidate gene approach by analyzing the genes encoding the PPR proteins that are located in the vicinity of the QTLs. In this group, genes that encode PPR proteins predicted to be targeted to mitochondria and belonging to the PLS subfamily, specific to plants, are the most promising candidates. Mapping editing efficiency polymorphisms may facilitate the identification of the factors controlling RNA editing in plant mitochondria, leading to a greater understanding of the mechanism of action of genes that regulate this intriguing posttranscriptional process.

	MATERIALS AND METHODS

TOP ABSTRACT RESULTS DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

 

Plant Material

Seeds of Arabidopsis (Arabidopsis thaliana) ecotype Col-4 (N933), Ler-0 (NW20), and the RILs were obtained from the Nottingham Arabidopsis Stock Centre (NASC; http://nasc.nott.ac.uk). Seeds were sown in Metromix soil and transferred in the dark at 4°C for 5 d before growing in a growth chamber at 22°C in an 8-h-dark/16-h-light cycle.

Molecular Biology Techniques

Total RNA was extracted using a RNeasy Plant mini kit (Qiagen) and treated with a DNA-free kit (Ambion). First-strand cDNA was synthesized from 1.5 µg of DNA-free RNA for 1 h at 37°C with an Omniscript kit (Qiagen) using the gene-specific reverse primer At-R (Table V) following the manufacturer's protocol. Reactions without reverse transcriptase were performed to check for genomic DNA contamination whenever the transcript did not contain an intron. cDNA samples were amplified by a standard protocol (5 min at 94°C followed by 35 cycles of 94°C, 30 s, 50°C, 30 s, 72°C, 2 min) in a PTC-200 thermal cycler (MJ Research) using gene-specific forward and reverse primers At-F and At-R (Table V).

Genomic DNA was extracted (Dellaporta et al., 1983) and cleaved amplified polymorphic sequence analysis of F₁ hybrids was performed (Konieczny and Ausubel, 1993) using alcohol dehydrogenase primers.

PPE of RT-PCR products and determination of editing efficiency were conducted as previously described (Peeters and Hanson, 2002) using the primers PPE-gene-site (Table V). To confirm the results, all experiments were performed three times.

Statistical Analysis

The estimation of the variance of the editing efficiency between the RILs and within the RILs was done by fitting a variance component model using the mixed procedure in the SPSS software version 13.0 for PC. A likelihood ratio statistic test performed by using SPSS allowed the determination as to whether the variance between the RILs was significantly different from zero.

The chromosomal location of QTLs for editing efficiency was determined by three analytical methods, SPA using the Qgene computer program version 3.07 (Nelson, 1997), IM, and CIM, using Windows QTL Cartographer version 2.5 (Wang et al. 2005). CIM was implemented by using the standard model 6 (as recommended by the authors of the software as default) specifying five cofactors to control for genetic background and a window size of 10 cM that blocked out a region of the genome on either side of the markers flanking the test site. Since the flanking regions are tightly linked to the test site, they have to be eliminated from the background markers in order to be able to detect a signal from the test site itself. The specific cofactors used in the model 6 were obtained by two different methods, a forward regression and a forward-backward stepwise regression with P_in = 0.01 and P_out = 0.01. Because the QTLs identified with these two methods are the same, only the data obtained with the forward regression will be presented. The walking speed for both CIM and IM was set up to 2 cM.

We used permutation tests to establish experiment-wise significance thresholds for the three different analyses, as suggested by Churchill and Doerge (1994). For SPA, 10,000 permutations allowed the determination of a threshold of F > 13.66 or LOD > 2.59 that corresponds to an experiment-wise threshold of P < 0.05. An experiment-wise threshold of P < 0.05 was also chosen for the IM and the CIM and corresponds to LODs > 2.7 for both analyses (1,000 permutations). A more stringent experiment-wise threshold of P < 0.01 was computed in the same way for the three analyses and corresponds to LOD > 3.35 (SPA), LOD > 4 (IM), and LOD > 3.8 (CIM).

The proportion of observed phenotypic variance attributable to a particular QTL was estimated by the coefficient of determination (R²) from the corresponding linear model (SPA) and using maximum likelihood for IM and CIM.

Pairwise epistatic interactions between QTLs were tested by a two-way ANOVA using Qgene software. Multiple regression analysis was also conducted by using Qgene software.

The search for mitochondrial editing sites homologous to ccb206 C24 was performed by using the Pratt computer program (Jonassen, 1997). Pratt software was developed to identify a pattern in a DNA sequence. The pattern may contain a flexible number of gaps and ambiguous symbols at certain positions. The user supplies a set of unaligned sequences and the minimum number of sequences to match the pattern. During the first phase, patterns are constrained to the pattern class defined by the set of options given by the user. The pattern graph is constructed using a depth-first search algorithm. A search is done for the highest scoring patterns in the class that is derived from the pattern graph. The highest scoring patterns found during this search are input to a heuristic pattern refinement algorithm, where more ambiguous pattern components can be used. In case flexible gaps are allowed, it is not guaranteed that the heuristic optimization algorithm will find the highest scoring pattern. The program analyzed 456 sequence segments of length 20 containing the 20 nucleotides upstream of each reported RNA-editing site (Giege and Brennicke, 1999). The input parameters that were used are as follows: CM = 5, C% = 1.1, pp = start, PL = 20, PN = 20, PX = 3, FN = 4, FL = 3, FP = 10, BI = off, BN = 20, S = info, G = query, GF = seq.fasta, E = 2, R = on, RG = off, OP = on, ON = 200, OA = 200, M = on, MR = on, MV = on. The input format can be found at http://us.expasy.org/tools/pratt/pratt-doc.html%23how1. The file seq.fasta contains a segment of the sequence ccb206 C24 ggagcctggtcttgactcttc. Highly conserved patterns with high scores were chosen as candidates for further screening.

	ACKNOWLEDGMENTS

 
We thank Dr. Tamara Galor from the Theory Center at Cornell University for performing the search for sites similar to ccb206 C24 with the Pratt software. We wish to express our thanks to Susan McCouch and Clare Nelson for their helpful comments, and to Françoise Vermeylen from the office of Statistical Consulting at Cornell University for her help with the statistical analyses.

Received July 27, 2005; returned for revision September 13, 2005; accepted September 23, 2005.

	FOOTNOTES

 
¹ This work was supported by grants from the National Institutes of Health (R01 GM50723) and the National Science Foundation (MCB 0344007).

² These authors contributed equally to the paper.

³ Present address: Unité Mixte de Recherche 990, Institut National de la Recherche Agronomique (INRA)/Institut National Polytechnique-Ecole Nationale Supérieure Agronomique, "Genomique et Biotechnologie des Fruits," Centre INRA Toulouse Auzeville, Chemin de Borde Rouge, 31320 Castanet-Tolosan, France.

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Stéphane Bentolila (sb46{at}cornell.edu).

Article, publication date, and citation information can be found at www.plantphysiol.org/cgi/doi/10.1104/pp.105.069013.

^* Corresponding author; e-mail mrh5{at}cornell.edu; fax 607–255–6249.

	LITERATURE CITED

TOP ABSTRACT RESULTS DISCUSSION CONCLUSION MATERIALS AND METHODS LITERATURE CITED

 
Bentolila S, Alfonso AA, Hanson MR (2002) A pentatricopeptide repeat-containing gene restores fertility to cytoplasmic male-sterile plants. Proc Natl Acad Sci USA 99: 10887–10892[Abstract/Free Full Text]

Blanc V, Navaratnam N, Henderson JO, Anant S, Kennedy S, Jarmuz A, Scott J, Davidson NO (2001) Identification of GRY-RBP as an apolipoprotein B RNA-binding protein that interacts with both apobec-1 and apobec-1 complementation factor to modulate C to U editing. J Biol Chem 276: 10272–10283[Abstract/Free Full Text]

Brown GG, Formanova N, Jin H, Wargachuk R, Dendy C, Patil P, Laforest M, Zhang J, Cheung WY, Landry BS (2003) The radish Rfo restorer gene of Ogura cytoplasmic male sterility encodes a protein with multiple pentatricopeptide repeats. Plant J 35: 262–272[CrossRef][Web of Science][Medline]

Chateigner-Boutin AL, Hanson MR (2002) Cross-competition in transgenic chloroplasts expressing single editing sites reveals shared cis elements. Mol Cell Biol 22: 8448–8456[Abstract/Free Full Text]

Chateigner-Boutin AL, Hanson MR (2003) Developmental co-variation of RNA editing extent of plastid editing sites exhibiting similar cis-elements. Nucleic Acids Res 31: 2586–2594[Abstract/Free Full Text]

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[Web of Science][Medline]

Choury D, Farre JC, Jordana X, Araya A (2004) Different patterns in the recognition of editing sites in plant mitochondria. Nucleic Acids Res 32: 6397–6406[Abstract/Free Full Text]

Churchill G, Doerge RW (1994) Empirical threshold values for quantitative trait mapping. Genetics 138: 963–971[Abstract]

Dellaporta SL, Wood J, Hicks JB (1983) A plant DNA minipreparation: version II. Plant Mol Biol Rep 1: 19

Desloire S, Gherbi H, Laloui W, Marhadour S, Clouet V, Cattolico L, Falentin C, Giancola S, Renard M, Budar F, et al (2003) Identification of the fertility restoration locus, Rfo, in radish, as a member of the pentatricopeptide-repeat protein family. EMBO Rep 4: 588–594[CrossRef][Web of Science][Medline]

Emanuelsson O, Nielsen H, Brunak S, von Heijne G (2000) Predicting subcellular localization of proteins based on their N-terminal amino acid sequence. J Mol Biol 300: 1005–1016[CrossRef][Web of Science][Medline]

Farré JC, Leon G, Jordana X, Araya A (2001) Cis recognition elements in plant mitochondrion RNA editing. Mol Cell Biol 21: 6731–6737[Abstract/Free Full Text]

Giege P, Brennicke A (1999) RNA editing in Arabidopsis mitochondria affects 441 C to U changes in ORFs. Proc Natl Acad Sci USA 96: 15324–15329[Abstract/Free Full Text]

Gott JM, Emeson RB (2000) Functions and mechanisms of RNA editing. Annu Rev Genet 34: 499–531[CrossRef][Web of Science][Medline]

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[Abstract/Free Full Text]

Hanson MR, Bentolila S (2004) Interactions of mitochondrial and nuclear genes that affect male gametophyte development. Plant Cell (Suppl) 16: S154–S169

Hashimoto M, Endo T, Peltier G, Tasaka M, Shikanai T (2003) A nucleus-encoded factor, CRR2, is essential for the expression of chloroplast ndhB in Arabidopsis. Plant J 36: 541–549[CrossRef][Web of Science][Medline]

Hegeman CE, Hayes ML, Hanson MR (2005) Substrate and cofactor requirements for RNA editing of chloroplast transcripts in Arabidopsis in vitro. Plant J 42: 124–132[CrossRef][Web of Science][Medline]

Hirose T, Sugiura M (2001) Involvement of a site-specific trans-acting factor and a common RNA-binding protein in the editing of chloroplast mRNAs: development of a chloroplast in vitro RNA editing system. EMBO J 20: 1144–1152[CrossRef][Web of Science][Medline]

Itani K, Handa H (1998) Rapeseed mitochondrial ccb206, a gene involved in cytochrome c biogenesis, is co-transcribed with the nad3 and rps12 genes: organization, transcription, and RNA editing of the nad3/rps12/ccb206 locus. Curr Genet 34: 318–325[CrossRef][Web of Science][Medline]

Jonassen I (1997) Efficient discovery of conserved patterns using a pattern graph. Comput Appl Biosci 13: 509–522[Abstract/Free Full Text]

Karcher D, Bock R (1998) Site-selective inhibition of plastid RNA editing by heat shock and antibiotics: a role for plastid translation in RNA editing. Nucleic Acids Res 26: 1185–1190[Abstract/Free Full Text]

Keegan LP, Gallo A, O'Connell MA (2001) The many roles of an RNA editor. Nat Rev Genet 2: 869–878[Web of Science][Medline]

Koizuka N, Imai R, Fujimoto H, Hayakawa T, Kimura Y, Kohno-Murase J, Sakai T, Kawasaki S, Imamura J (2003) Genetic characterization of a pentatricopeptide repeat protein gene, orf687, that restores fertility in the cytoplasmic male-sterile Kosena radish. Plant J 34: 407–415[CrossRef][Web of Science][Medline]

Komori T, Ohta S, Murai N, Takakura Y, Kuraya Y, Suzuki S, Hiei Y, Imaseki H, Nitta N (2004) Map-based cloning of a fertility restorer gene, Rf-1, in rice (Oryza sativa L.). Plant J 37: 315–325[CrossRef][Web of Science][Medline]

Konieczny A, Ausubel FM (1993) A procedure for mapping Arabidopsis mutations using co-dominant ecotype-specific PCR-based markers. Plant J 4: 403–410[CrossRef][Web of Science][Medline]

Kotera E, Tasaka M, Shikanai T (2005) A pentatricopeptide repeat protein is essential for RNA editing in chloroplasts. Nature 433: 326–330[CrossRef][Medline]

Kurihara-Yonemoto S, Handa H (2001) Low temperature affects the processing pattern and RNA editing status of the mitochondrial cox2 transcripts in wheat. Curr Genet 40: 203–208[CrossRef][Web of Science][Medline]

Lister C, Dean C (1993) Recombinant inbred lines for mapping RFLP and phenotypic markers in Arabidopsis markers in Arabidopsis thaliana. Plant J 4: 745–750[CrossRef][Web of Science]

Lu B, Hanson MR (1992) A single nuclear gene specifies the abundance and extent of RNA editing of a plant mitochondrial transcript. Nucleic Acids Res 20: 5699–5703[Abstract/Free Full Text]

Lurin C, Andres C, Aubourg S, Bellaoui M, Bitton F, Bruyere C, Caboche M, Debast C, Gualberto J, Hoffmann B, et al (2004) Genome-wide analysis of Arabidopsis pentatricopeptide repeat proteins reveals their essential role in organelle biogenesis. Plant Cell 16: 2089–2103[Abstract/Free Full Text]

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[Abstract/Free Full Text]

Nakajima Y, Mulligan RM (2001) Heat stress results in incomplete C-to-U editing of maize chloroplast mRNAs and correlates with changes in chloroplast transcription rate. Curr Genet 40: 209–213[CrossRef][Web of Science][Medline]

Nelson JC (1997) QGENE: software for marker-based genomic analysis and breeding. Mol Breed 3: 239–245

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 sequence acquisition and loss during the evolution of flowering plants. Mol Genet Genomics 268: 434–445[CrossRef][Web of Science][Medline]

Peeters NM, Hanson MR (2002) Transcript abundance supercedes editing efficiency as a factor in developmental variation of chloroplast gene expression. RNA 8: 497–511[Abstract]

Sambrook J, Russell DW (2001) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

Schuster W (1994) The highly edited orf206 in Oenothera mitochondria may encode a component of a heme transporter involved in cytochrome c biogenesis. Plant Mol Biol 25: 33–42[CrossRef][Medline]

Small I, Peeters N, Legeai F, Lurin C (2004) Predotar: a tool for rapidly screening proteomes for N-terminal targeting sequences. Proteomics 4: 1581–1590[CrossRef][Web of Science][Medline]

Small ID, Peeters N (2000) The PPR motif—a TPR-related motif prevalent in plant organellar proteins. Trends Biochem Sci 25: 46–47[Web of Science][Medline]

Stern DB, Hanson MR, Barkan A (2004) Genetics and genomics of chloroplast biogenesis: maize as a model system. Trends Plant Sci 9: 293–301[CrossRef][Web of Science][Medline]

Sutton CA, Zoubenko OV, Hanson MR, Maliga P (1995) A plant mitochondrial sequence transcribed in transgenic tobacco chloroplasts is not edited. Mol Cell Biol 15: 1377–1381[Abstract/Free Full Text]

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[Abstract/Free Full Text]

Takenaka M, Neuwirt J, Brennicke A (2004) Complex cis-elements determine an RNA editing site in pea mitochondria. Nucleic Acids Res 32: 4137–4144[Abstract/Free Full Text]

Tsudzuki T, Wakasugi T, Sugiura M (2001) Comparative analysis of RNA editing sites in higher plant chloroplasts. J Mol Evol 53: 327–332[CrossRef][Web of Science][Medline]

Wang S, Basten CJ, Zeng ZB (2005) Windows QTL Cartographer 2.5. Department of Statistics, North Carolina State University, Raleigh, NC

This article has been cited by other articles:

				J. C. Robbins, W. P. Heller, and M. R. Hanson A comparative genomics approach identifies a PPR-DYW protein that is essential for C-to-U editing of the Arabidopsis chloroplast accD transcript RNA, June 1, 2009; 15(6): 1142 - 1153. [Abstract] [Full Text] [PDF]

				J. Forner, A. Holzle, C. Jonietz, S. Thuss, M. Schwarzlander, B. Weber, R. C. Meyer, and S. Binder Mitochondrial mRNA Polymorphisms in Different Arabidopsis Accessions Plant Physiology, October 1, 2008; 148(2): 1106 - 1116. [Abstract] [Full Text] [PDF]

				M. Rudinger, M. Polsakiewicz, and V. Knoop Organellar RNA Editing and Plant-Specific Extensions of Pentatricopeptide Repeat Proteins in Jungermanniid but not in Marchantiid Liverworts Mol. Biol. Evol., July 1, 2008; 25(7): 1405 - 1414. [Abstract] [Full Text] [PDF]

				S. Bentolila, L. E. Elliott, and M. R. Hanson Genetic Architecture of Mitochondrial Editing in Arabidopsis thaliana Genetics, March 1, 2008; 178(3): 1693 - 1708. [Abstract] [Full Text] [PDF]

				A.-L. Chateigner-Boutin and I. Small A rapid high-throughput method for the detection and quantification of RNA editing based on high-resolution melting of amplicons Nucleic Acids Res., September 27, 2007; 35(17): e114 - e114. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 139/4/2006    most recent pp.105.069013v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Web of Science (12) Citing Articles via Google Scholar Google Scholar Articles by Bentolila, S. Articles by Hanson, M. R. Search for Related Content PubMed PubMed Citation Articles by Bentolila, S. Articles by Hanson, M. R. Agricola Articles by Bentolila, S. Articles by Hanson, M. R.