SEARCH

SEARCH BY CITATION

Keywords:

  • mitochondrial gene expression;
  • nuclear–mitochondrial interactions;
  • RNA stability;
  • RNA processing;
  • species-specific factors

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary Material
  9. References
  10. Supporting Information

Mitochondrial transcription was investigated in a cytoplasmic male-sterile (CMS) Brassica napus line with rearranged mitochondrial (mt) DNA mostly inherited from Arabidopsis thaliana. The transcript patterns were compared with the corresponding male-fertile progenitors, B. napus and A. thaliana, and a fertility-restored line. Transcriptional activities, gene stoichiometry and transcript steady-state levels were analysed for all protein and rRNA coding genes and for several orfs present in the A. thaliana mitochondrial genome. The transcriptional activities were highly variable when comparing the parental species, while the CMS and restored lines displayed similar activities. For several ribosomal protein genes transcriptional activity was reduced while it was increased for orf139 in comparison with the parental species. The differences in transcriptional activity observed could be related to differences in relative promoter strength, as gene stoichiometry between lines was very limited. Transcript steady-state levels were more homogenous than the transcriptional activities demonstrating RNA turnover as a compensating mechanism. In the CMS line higher transcript abundance and novel transcript patterns in comparison with the parental lines were found for several genes. Of those, the transcripts for orf139, orf240a and orf294 were less abundant in the fertility-restored line. These putative CMS-associated transcripts were mapped by cRT-PCR. In conclusion we show that (mt) DNA from A. thaliana was non-correctly transcribed and processed/degraded in the B. napus nuclear background. Furthermore, the introgressed nuclear A. thaliana DNA in the fertility-restored line contributes to a more rapid degradation of transcripts accumulated from A. thaliana derived orfs in the CMS line.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary Material
  9. References
  10. Supporting Information

Plant mitochondrial gene expression is regulated in a complex way, most likely by nuclear control of a cascade of transcriptional and post-transcriptional events (reviewed by Binder and Brennicke, 2003; Giegé and Brennicke, 2001). These events determine the final steady-state population of translatable RNA. Even though plant mitochondrial functions are similar in most species, the composition of the mitochondrial DNA (mtDNA) in size, structure and sequence is highly variable (Gray, 1999). How the nuclearly encoded proteins regulate gene expression from these variable genomes is rather unknown. Thus, to gain more knowledge about species-specific regulation of plant mitochondrial gene expression, we have compared different steps of transcription in an alloplasmic plant material with various nuclear–mitochondrial compositions.

Plant mitochondria, in contrast to animal and yeast mitochondria, have multiple promoters of different strength (Brennicke et al., 1999; Fey and Maréchal-Drouard, 1999). Sometimes several promoters act on single genes and in other cases one promoter initiates transcription of several genes (Binder et al., 1996). The completion of the Arabidopsis thaliana mtDNA sequence (Unseld et al., 1997) allowed a search for promoter sites containing a conserved nanonucleotide promoter element within this genome (Dombrowski et al., 1998). As many genes lack this conserved type of promoter, studies (Dombrowski et al., 1998; Fey and Maréchal-Drouard, 1999; Remacle and Maréchal-Drouard, 1996) suggest that additional non-conserved promoter sequences must be active in plant mitochondria. These cryptic promoters can be species-specific, as shown in alloplasmic tobacco (Edqvist and Bergman, 2002) and maize (Newton et al., 1995). In addition to promoter type, transcriptional rates can also be regulated by selective gene amplification or suppression (Janska et al., 1998; Muise and Hauswirth, 1995). Equally important for the final amount of translatable RNA are post-transcriptional processes (Finnegan and Brown, 1990; Giegéet al., 2000; Mulligan et al., 1991). These include splicing, editing, 5′ and 3′ end maturation and regulation of transcript degradation. Similar to promoter elements several conserved internal sequences for splicing, editing and maturation have been identified (Binder and Brennicke, 2003). Experiments with the atp9 gene in potato and pea (Gagliardi et al., 2001; Kuhn et al., 2001) have revealed that stabilizing stem-loop structures and destabilizing polyadenylation of 3′ ends can determine degradation rates of transcripts.

Protoplast fusion is an effective means of altering mtDNA by combining the cytoplasms of two genotypes in one cell, usually resulting in rearrangements in the mitochondrial genome (Earle, 1995; Pelletier, 1986). Furthermore, hybrids produced between distantly related species often lead to alloplasmic incompatibilities between nuclear and mitochondrial genes. This can result in cytoplasmic male-sterile (CMS) plants that are unable to produce functional anthers and/or pollen. In most CMS systems the trait has been associated with transcription and translation of novel open reading frames (orfs) (Hanson and Bentolila, 2004; Schnable and Wise, 1998). Specific nuclear genes, often introgressed from the cytoplasmic donor species, have the ability to repress the expression of the CMS-associated orfs and restore male fertility. These genes, termed restorer of fertility (Rf) genes, act on different transcriptional mechanisms, including regulation of gene copy number abundance (Bellaoui et al., 1998; Janska et al., 1998), transcript initiation (Edqvist and Bergman, 2002), transcript maturation (Dill et al., 1997; Singh et al., 1996) and transcript stability (Gagliardi and Leaver, 1999).

In the present study we have screened for species-specific steps in mitochondrial gene expression and investigated whether transcriptional aberrations could be counter-balanced by post-transcriptional events. By the assessment of different techniques, including run-on transcription, kinase labelling of transcripts and DNA as well as RFLP and Northern analysis and with multiple gene probes, nuclear–mitochondrial interactions regulating transcriptional events were investigated on a genome-wide level. The plant material chosen is a CMS system obtained from protoplast fusions between Arabidopsis thaliana and Brassica napus (Leino et al., 2003). The availability of the complete mitochondrial genome sequence of both these species (Handa, 2003; Unseld et al., 1997), and the large difference in structure and size between them, makes the material especially suitable for studies of species-specific effects on mitochondrial transcriptional regulation. The CMS line contains mtDNA mainly from A. thaliana with some mtDNA from B. napus, whereas the nucleus contains pure B. napus DNA. This line displays complete male sterility and homeotic conversions of stamens to carpelloid structures. The fertility-restored line is isogenic with respect to the mitochondrial genome, but has the addition of one pair of A. thaliana chromosome III in the nuclear genome and has a phenotype reverted to male fertility (Leino et al., 2004). The results show a high degree of species-specific mitochondrial transcriptional regulation and alloplasmic-induced aberrations.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary Material
  9. References
  10. Supporting Information

Run-on assays reveal diverse transcriptional rates

Run-on assays were performed in order to measure transcriptional activity of mitochondrial genes in the CMS and fertility-restored lines as well as in the fertile parental species. Crude mitochondrial preparations from flower buds were disrupted in the presence of [α-32P]G/UTP to extend previously initiated transcripts. Transcripts thus labelled were used as probes on membrane-bound mtDNA sequences (Figure 1a–d). These sequences represent all identified protein coding, rRNA and conserved unidentified genes in the A. thaliana mt-genome. Additionally, sequences for several orfs in the A. thaliana mt-genome, previously found expressed (Giegéet al., 2000) or with mosaic structures (Marienfeld et al., 1997) were included on the membranes. Control hybridizations (not shown) of dilution series of filter-bound target DNA verified that the target molecules on the filter were in excess and not saturated by the probe. To remove unannealed RNA the filters were RNase treated prior to exposure. Control experiments with in vitro transcribed RNA of defined lengths showed that this treatment efficiently removed all single-stranded non-annealed RNA (data not shown). The hybridization signal for each gene was quantified by phosphorimaging and the data were normalized for the amount of UTP or GTP incorporated in each transcript based on the coding-strand sequence of the target DNA. The values for each gene were expressed as a percentage of the total hybridization signal generated by each probe, thus indicating the relative transcriptional activity.

image

Figure 1. Phosphoimages of membranes containing all identified Arabidopsis mitochondrial protein and rRNA coding genes and several orfs hybridized with (a)–(d): run-on transcripts or (f)–(i): end-labelled steady-state transcripts from the lines. (a), (f): Brassica napus; (b), (g): Arabidopsis thaliana; (c), (h): CMS; (d), (i): fertility restored. (e) Hybridization with a vector probe to normalize for target DNA abundance. (j) Gene order on membranes.

Download figure to PowerPoint

Figure 2(a) illustrates transcriptional activity, i.e. the relative run-on signal, for the mitochondrial genes. Detailed data are found online in Table S2. The rates varied among genes over a 20-fold range in all lines. Overall, the rRNA genes were the most actively transcribed, whereas many of the unidentified orfs were the least transcribed. Interestingly, F-tests showed significant differences among the lines for most genes. These genes were subjected to multiple pairwise comparisons (Table S2). To avoid the risk for type I comparison-wise errors, significance levels were Bonferroni corrected (P < 0.00022). The parental lines A. thaliana and B. napus differed significantly for the genes atp8, ccmB, rps7 and rrn5 which had higher transcriptional activity in B. napus than in A. thaliana. The opposite relationship was found for the genes nad4L, nad9 and cox1. For most of the A. thaliana-derived orfs the signal in B. napus was close to zero. In contrast to the many differences found between the two parental species, the values obtained for the CMS and fertility-restored line are very similar (Figure 2a). Although the rates of transcription for most genes are in the same range as for either of the parental lines, certain genes had higher or lower transcriptional activity than that found for both B. napus and A. thaliana. In particular, significantly lower rates of the ribosomal protein genes, rpl5, rpl16, rps3 and rps12 were observed. The most apparent difference was found for orf139. This gene was expressed in the same rates as the rrn-genes and had fivefold higher transcriptional activity in the CMS line than in A. thaliana (Figure 3h, section 1) and 50% higher activity than in the restored line.

image

Figure 2. Relative transcriptional rates and transcript abundance of mitochondrial genes in the different lines. The values are averages of four independent experiments and are expressed as percentage of the total amount observed in each line respectively. Genes with significant differences between lines (F-test, P < 0.05) are indicated in bold. B, Brassica napus; A, Arabidopsis thaliana; C, CMS; R, fertility restored. A scale is shown at bottom. (a) Transcriptional rates as measured by run-on assays. (b) Transcript abundance as measured by kinase end-labelling of steady-state transcripts.

Download figure to PowerPoint

image

Figure 3. Comparative analysis of (1) transcriptional activities, (2) mt-RFLP patterns, (3) transcript steady-state levels and (4) Northern gel-blot analysis for specific genes. Values in (1) and (3) are expressed as a percentage of the total values from each line, the streaks show the 1% levels, except for (h) where the streaks show the 2% levels. Error bars show standard deviation. Lines are: B, Brassica napus; A, Arabidopsis thaliana; C, CMS; R, fertility restored. Genes are: (a), atp1; (b), atp9; (c), ccmC; (d), cox1; (e), cox2; (f), cox3; (g), orf240a; (h), orf139; (i), orf294.

Download figure to PowerPoint

Gene stoichiometry and mtDNA composition

As selective amplification of certain sequences could give rise to variable transcriptional rates, we assayed for possible differential mtDNA sequence abundance. Fragments obtained from sonicated mtDNA were kinase end-labelled and used as probes on the same filters as described above. The hybridization signal was quantified and the sequence abundance estimated. In comparison with the range of variation observed in transcriptional rates, only minor differences were found between the lines (Table S3). Between the two parental species hybridization signal intensity differed more than twofold for only two genes, i.e. nad4 (2.0) and cox2 (2.1). The cox2 gene is duplicated in the B. napus mt-genome (Handa, 2003), but not in A. thaliana. Furthermore, in B. napus the deduced sequence abundance for many of the unidentified A. thaliana orfs is low. These genes have mosaic structures, with sequences partly present in the B. napus mt-genome, which probably causes some hybridization. In the CMS and restored line relative signal intensities are similar and in neither case do they exceed a twofold difference in comparison with the parental lines. The signals from all of the unidentified A. thaliana orfs were similar to those of A. thaliana in both the CMS and restored line.

To verify the stoichiometric analysis and the parental origin of the mt-genes in the CMS/restored line, RFLP analysis was performed using the same probes as used in the run-on assay (Figure 3, sections 2). The results (Table S4) confirm those of Leino et al. (2003) showing that the larger part of the mt-genome in the CMS/restored line not only originates from A. thaliana, but also that it contains some sequences of B. napus origin. Figure 4 illustrates the position of the genes, investigated in this study, in the A. thaliana and B. napus genome. By using the RFLP data the gene regions derived from each parental species in the CMS/restored line is visualized. Some genes are inherited from both parents, i.e. nad9, atp6, rpl2, orfX, ccmFN2 (not shown) and ccmC (Figure 3c, section 2). Correspondingly, the relative sequence abundance was found to be higher for these genes (Table S3). Rearranged patterns with novel non-parental fragments are found with the probes nad1a, rps4 (not shown), atp1 and cox2 (Figure 3a,e, section 2). For nad1a, atp1 and cox2 higher relative sequence abundance was also observed although not exceeding twofold in comparison with the parental species (Table S3). A different pattern was found in the restored line in comparison with the CMS line when hybridizing with the cox2 gene probe, indicating nuclear influence on the mt-genome structure. With the atp1 probe the same RFLP pattern was found in the CMS and restored line but with strong differences in hybridization signal for the different fragments.

image

Figure 4. Gene organization of the Arabidopsis thaliana and Brassica napus mitochondrial genomes based on Unseld et al. (1997) and Handa (2003). Reading frames transcribed from the clockwise strand are indicated at the outer side, while reading frames transcribed from the counter-clockwise strand are indicated on the inner side. Black boxes (bsl00001) represent genes present in the CMS/restored line from each parental species, as determined by RFLP analysis. Pale grey boxes ( inline image) represent genes not present in the CMS/restored line. Dark grey boxes ( inline image) indicate genes not determined or indeterminate in the RFLP analysis.

Download figure to PowerPoint

Hybridizations with end-labelled RNA show balanced transcript steady-state levels

The steady-state population of mtRNA molecules was analysed and compared with transcriptional activity in order to estimate the differences in RNA turnover. Isolated RNA was end-labelled enzymatically, making signal intensity independent of transcript length. As, in some cases, steady-state transcripts do not undergo 5′ processing (Giegé and Brennicke, 2001), the polynucleotide kinase labelling was preceded by a dephosphorylation step that enables 5′ processed as well as the de novo transcripts to be labelled. Labelled transcripts were then used as probes on the same membrane-bound mt gene sequences (Figure 1f–i) as used in the run-on experiments. Signal intensities were calculated as a percentage of the total signal obtained from all target genes. The relative steady-state levels are illustrated in Figure 2b with detailed data available online in Table S5.

In general, transcripts for the genes of the respiratory chain subunits were more abundant than genes for cytochrome c biosynthesis, ribosomal proteins and unidentified genes. The rRNA genes constitute the bulk of steady-state RNA. Surprisingly low levels of rrn5 transcripts were found. Mature rrn5 transcripts are much shorter than the other orfs/genes investigated in this study and could have been lost during the purification of the probe. To test this we ran RNA size markers in the same purification steps and found that transcripts shorter than 200 nucleotides are sorted out to a large extent (data not shown). Thus, we believe that the abundance of the 119 nt rrn5 transcripts probably is underestimated. Transcript abundance among lines was balanced for more genes in comparison with transcriptional activity. Although significant F-tests were found for several genes (Figure 2), no significant differences between the parental lines were found in the pairwise comparisons (Bonferroni correction, P < 0.00043). In the CMS line significantly higher transcript abundance was found for the genes cox3/sdh4, atp9, ccmFC and orf240a in comparison with the parental species. The transcript abundance was reduced for these genes in the restored line, although not to the parental species level (Figures 2 and 3b,f,g section 3). For many transcripts with low steady-state levels, i.e. signals only slightly above background, the differences between lines are more difficult to distinguish. Small differences in the background cut-off have large impact on the relative levels between the lines. Although the background setting might be slightly underestimated, this minimizes the risk of finding false differences between lines. In particular, many of the unidentified A. thaliana orfs had very low steady-state values. Notably, orf139, with strong transcriptional activity in the CMS and restored line, had almost undetectable steady-state levels.

Northern analysis reveals presence of additional transcripts in the CMS line

To verify the quantitative levels estimated by the kinase labelling experiment, especially for low-abundant transcripts, and to determine qualitative alterations of transcripts, Northern analysis was performed with the same gene probes and RNA as used in the dot-blot analysis. For most probes complex transcription patterns with several transcripts were found (Figure 3, sections 4). In only a few cases did the two parental lines show similar transcription patterns, e.g. for atp9, cox1 (Figure 3b,d, section 4) and ccmFC (not shown). For most gene probes the CMS and restored line displayed transcription patterns similar to either or both parental lines. For several gene probes additional transcripts were present. Comparisons with RFLP data (Figure 4, Table S4) demonstrate that presence of B. napus mtDNA resulted in B. napus-like transcription patterns (not shown), while mtDNA from A. thaliana often resulted in long additional transcripts not present in either of the parental lines, e.g. with nad9, rpl16 (not shown), ccmC (Figure 3c, section 4), cox3 (Figure 3f, section 4) and atp9 (Figure 3b, section 4). In the restored line the amounts of the additional transcripts appeared to be smaller than in the CMS line. orf139 (Figure 3g, section 4), orf240a (Figure 3h, section 4) and orf294 (Figure 3i, section 4), giving weak signals in the kinase end-labelling assay, displayed clear differences between the lines in the Northern gel blot analysis. In all three cases a relatively abundant transcript was found in the CMS line in comparison with the restored line. In A. thaliana low level of transcripts was found using probe orf240a or orf294 but no transcripts were seen using the probe orf139. No transcripts were found in B. napus using either orf139, orf240a or orf294.

Mapping of transcript extremities of orf139, orf240a and orf294

The transcripts detected with the orf139, orf240a and orf294 probes, that showed large differences between the CMS and restored line, were examined in greater detail. Transcript extremities were mapped using a circular RT-PCR (cRT-PCR) strategy. RNA from A. thaliana, the CMS and restored line was self-ligated using T4 RNA ligase followed by cDNA synthesis with a specific primer and PCR, amplifying the sequence region containing the junction of 5′ and 3′ ends. PCR products obtained for the different gene regions and lines (Figure 5a) were cloned and sequenced. No PCR fragments were obtained in the absence of either the T4 RNA ligase or the reverse transcriptase (data not shown). Frequency, position and extent of the transcripts that correspond to the junctions of transcript ends are shown in Figure 5(b).

image

Figure 5. cRT-PCR analysis of orf139, orf240a and orf294 transcripts. (a) EtBr-stained agarose gel of obtained cRT-PCR products. M, size marker with fragment lengths in nt indicated on the left; A, Arabidopsis thaliana; C, CMS; R, fertility restored. (b) Schemes of the respective loci with direction of transcription indicated (bold arrow). Open reading frames (boxes) and location and orientation of primers for cDNA synthesis (arrows lettered c) and PCR (f, forward; r, reverse) are indicated. Locations of conserved nanonucleotide motifs (CNM) are marked. Frequency, position and extent of the transcripts corresponding to cRT-PCR clones are shown (thin streaks) in the respective lines, lettering is as in (a).

Download figure to PowerPoint

Amplified products were obtained from the CMS and restored line, but not from A. thaliana, using orf139-specific primers. A common 5′ end was found 1093 nt upstream of the start codon of orf139. Several different 3′ ends were found between 388 and 36 nt downstream of the stop codon. These ends correspond to transcripts between 1900 and 1548 nt, which is in accordance with the transcripts seen in the Northern analysis. In addition, fragments with shorter 5′ ends were also found. These shorter fragments were more frequent in the restored line. One clone from the CMS line corresponded to a 5′ end 1172 nt upstream of the start codon of orf139. Interestingly, this end is just 24 nt downstream of a conserved nanonucleotide motif (CNM). For orf240a all three lines had identical transcript ends (Figure 5b). The 5′ end mapped 30 nt upstream the start codon, just after the tRNA-Lys gene. The 3′ end mapped 199–201 nt downstream of the stop codon of orf240a, thus forming a 1000–1002 nt transcript. The cRT-PCR for orf294 resulted in products between 200 and approximately 1500 nt in all lines (Figure 5a). These products correspond to a wide range of different transcripts (Figure 5b). In general the A. thaliana transcripts were shorter than from the two other lines, mainly due to less extreme 5′ ends. The predominant 5′ ends mapped 15 or 32 nt upstream of the start codon of orf294. 3′ ends were heterogenous with the most extreme end 282 nt downstream of the stop codon. This 3′ part includes the first 136 nt of the atp1gene. More frequent ends were found at positions 272 nt and 172 nt downstream of the stop codon. In the CMS and restored line a common 5′ end was found 478 nt upstream of the start codon of orf294. 3′ ends were highly variable, though, especially in the restored line, common transcript ends were found either 272 nt downstream of the stop codon as in A. thaliana, or 9 nt downstream of the stop codon.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary Material
  9. References
  10. Supporting Information

Transcriptional and post-transcriptional control of gene regulation was investigated across the mitochondrial genome in an alloplasmic CMS line together with the male-fertile progenitors and a fertility-restored line. Our results show that transcriptional activity is primarily set by promoter strength which is species-specific and affected by nuclear background. Furthermore, it was demonstrated that RNA turnover counterbalances differential expression activity in the parental species. These processes were found less effective in the CMS line, indicating species-specific nuclear control of post-transcriptional events.

The diverse transcription rates, as estimated by the run-on assay, show that B. napus and A. thaliana have promoters of different strength for different genes, e.g. for cox1, nad4L, nad9, ccmB, rps7 and rrn5. Large differences in promoter strengths for specific genes among different species were also found in investigations performed in maize (Finnegan and Brown, 1990; Muise and Hauswirth, 1992; Mulligan et al., 1991), Brassica hirta (Muise and Hauswirth, 1995) and A. thaliana (Giegéet al., 2000). As shown by Unseld et al. (1997) and Handa (2003) many genes are closely positioned in the A. thaliana and B. napus mitochondrial genomes suggesting that these genes could be co-transcribed and regulated by the same promoter. For some gene clusters, e.g. the rps3-rpl16-nad9 in A. thaliana and rps4-nad2a in B. napus similar transcription rates were found. For other gene clusters, e.g. rps4-nad2a-nad6 in A. thaliana and rps3-rpl16-rpl5-cob in B. napus, very different run-on values were observed for individual genes within the cluster. These differences could be due to leaky transcription stops, or in the case of large increases in expression within a cluster, additional transcription starts. In spite of similar gene arrangement of the rrn18 and rrn5 genes in A. thaliana and B. napus, the two species have different relationships in transcription rates of the two genes. The intergenic sequence between rrn18 and rrn5 deviates between A. thaliana and B. napus, but whether it is this inconsistence and/or nuclear factors that cause the different patterns in transcription activity for rrn18/rrn5 for the two species is not clear. Transcript initiation in dicots has been correlated with a CNM (Binder et al., 1996). Searches for these elements in A. thaliana (Dombrowski et al., 1998) and B. napus (Handa, 2003) revealed that many genes lack this CNM in the 5′ UTRs and, thus, are under control of yet uncharacterized promoters. Furthermore, the identified CNMs are often located in other positions or absent for specific genes when comparing the homologous regions of B. napus and A. thaliana. Consequently, differences in promoter sequences can explain the very different transcriptional rates observed between the two species.

By comparing our data for A. thaliana and the data reported by Giegéet al. (2000), limited similarities concerning transcription activity were found even though the same run-on assay and mainly the same gene probes were used. For example, Giegéet al. (2000) found that the rRNA genes were transcribed in the same rates as the protein coding genes, whereas our results showed a 2- to 5-fold higher expression activity for the rRNA genes. Giegéet al. (2000) used mitochondria isolated from cell cultures whereas we used flower buds. These differences in materials might explain the divergences observed. Muise and Hauswirth (1992) reported on different transcriptional activity for individual genes when comparing different tissues from maize. This raises the question of whether plant mitochondrial promoters are differentially activated in different tissues. The CMS line and the restored line display morphological differences in floral development, as the anthers in the CMS line are homeotically converted to carpels. However, transcriptional rates are very similar in floral bud tissue from the CMS and restored line, indicating only minor differences in transcription activity of mitochondrial genes in different floral organs.

Comparisons of run-on values with the RFLP patterns show that the B. napus gene copies seem fully functional, whereas the A. thaliana gene copies influenced by the alloplasmic condition more often have an aberrant transcriptional activity. The ribosomal protein genes rpl5, rpl16, rps3 and rps12, all exclusively inherited from A. thaliana in the CMS line, show significantly lower expression activity. Dombrowski et al. (1998) did not find any CNM within a reasonable distance of any of these genes. Consequently, these genes are under transcriptional control of an uncharacterized promoter type. Our results indicate that these promoter elements are species-specific and thus dependent on the correct nuclear background for correct transcription. Orf139, which is expressed in the same rate as the rRNA genes in the CMS and restored lines, is preceded by a CNM. However, even though the most extreme 5′ end of the orf139 transcripts examined by cRT-PCR mapped only 24 nt downstream this motif, other promoters could be active. Interestingly, orf139 is present in repeats in the A. thaliana mt-genome and has been suggested to be active in intramolecular recombination (Unseld et al., 1997). A protoplast fusion-induced intergenomic recombination event at this site could have occurred and put the orf139 under control of a strong promoter.

The stoichiometric analysis excluded that variation in DNA sequence abundance could explain the different transcriptional activities observed. The same conclusion was made from studies in B. hirta, although gene copy numbers could be correlated with transcriptional activities in maize (Muise and Hauswirth, 1995). RFLP analysis using the cox2 gene as a probe revealed nuclear influence on the genome structure of the CMS line. The cox2 region is repeated in B. napus and possibly involved in the formation of the tricircular structure suggested for this species (Handa, 2003; Palmer and Herbon, 1988). In A. thaliana the nuclearly encoded CHM gene has been found to affect the mt-genome structure (Abdelnoor et al., 2003). This gene is positioned on chromosome III, which is present in the restored line in this study. Whether it is this gene or another that affects the rearranged mt-genome structure observed remains to be elucidated. The cox2 gene is situated 5 kb upstream of orf139a. Thus, the structural shifting observed might have influenced the promoter region for orf139 and the transcription activity of the orf, which differed significantly between the lines.

In comparison with run-on values it is obvious that transcript abundance does not directly correlate with transcriptional rates but is rather affected by RNA turnover. The extraordinary high stability of the rRNAs, as estimated by the differences between transcriptional rates and steady-state transcript levels, reported by Giegéet al. (2000), is confirmed in this study, although not as pronounced. The many differences in transcriptional activity between A. thaliana and B. napus found for specific genes are not reflected in the steady-state levels. Thus, the evenly balanced steady-state transcript levels must result from different rates of RNA degradation in the two species. In the CMS and restored line some of the alterations in expression seem to be buffered by selective enhanced or reduced degradation. For example, some of the ribosomal protein-coding genes that had reduced transcriptional activity have steady-state levels comparable to the parental lines. For some genes, transcript levels are significantly higher in the CMS line in comparison with the parental lines, e.g. cox3-sdh4, atp9 and ccmC. Northern analysis revealed that higher transcript abundance could be associated with additional non-parental transcripts. These transcripts could be explained either by the rearranged genome structures or, more probably, by nuclear influence as has been observed previously, e.g. in alloplasmic tobacco (Håkansson and Glimelius, 1991), maize (Wen and Chase, 1999) and Brassica (Li et al., 1998).

According to our results several A. thaliana transcripts accumulated in the CMS line, that by influence of the nuclear genes in the restored line were reduced. This was most obvious for the sequences orf139, orf240a and orf294. Orf139 displayed a very high transcriptional rate in the CMS and restored line, while it was degraded to very low steady-state levels. Surprisingly, the cRT-PCR experiments showed that the majority of 5′ ends map to a position over 1 kb upstream of the start codon. This position coincides with the last nucleotide of the small non-messenger RNA Ath-59 (Marker et al., 2002) that could serve as a processing site. A few clones with shorter 5′ extensions were also found. However, as shorter products are favoured in the cloning process, we assume these products to be rather rare. In contrast to the long 5′ extension, 3′ ends are much shorter and more heterogenous. Thus, we think that the 5′ end is relatively stable, at least in the CMS line, and that the transcripts are actively degraded from the 3′ end. Although the relationship in steady-state levels could be related to the difference in transcription rates, the inability to detect transcripts either with Northern analysis or cRT-PCR in A. thaliana suggests that degradation of this transcript is more efficient in A. thaliana.

The shift in transcription rates and steady-state levels for orf240a showed that the stability of these transcripts was different between the species. As the transcripts were identical in all three materials, i.e. A. thaliana, the CMS and restored line, we concluded that differential processing is not the cause of variable stability. A similar situation was observed in sunflower where introduction of the nuclear restorer gene resulted in a more rapid degradation of the CMS-associated orf522 transcripts promoted by polyadenylation of the 3′ ends (Gagliardi and Leaver, 1999). Although we did not find any polyadenylated orf240a transcripts in the cRT-PCR experiments, this may well occur. Only a limited number of clones were sequenced and as polyadenylated transcripts are possibly prone to rapid degradation polyA-tails could easily be missed by this technique. In addition, orf294 was more stable in the CMS line. Northern analysis revealed that the major transcript in the CMS and restored line was approximately 2500 nt and in A. thaliana approximately 2100 nt. The different 5′ ends found by cRT-PCR could explain the size difference. A common 5′ end in the CMS and restored line, mapped 478 nt upstream of the start codon, was not found in A. thaliana. Instead the predominant ends were found 15 or 32 nt upstream of the start codon. However, the resulting transcripts were too short to match the major transcript found by Northern analysis and we assumed that the mapped transcripts were degradation products with shortened 3′ ends. The cRT-PCR procedure does not favour the cloning of long transcripts, which could explain the failure to find full-length transcripts. We propose that the full-length transcripts extend further downstream in the atp1 gene. Indeed, Northern analysis with the atp1 probe displays transcripts of the same length and abundance as the orf294 probe. Additionally, the mature atp1 transcript appeared less abundant in the CMS line suggesting that the processing of the atp1 precursor transcripts is less efficient in the CMS line. Mapping of the 3′ ends should be performed with more 3′ extreme primers to find the true 3′ ends and possibly explain the difference in stability/processing of the transcripts.

Earlier investigations on plant mitochondrial gene expression in alloplasmic and/or CMS lines have been restricted to studies of a limited number of genes/orfs in most cases. In this study we have used probes for all identified and several uncharacterized protein coding genes in the A. thaliana mitochondrial genome. Many of the orfs have a mosaic structure with fragments from identified mitochondrial genes and additional sequences of unknown origin (Marienfeld et al., 1997). Others are positioned in close proximity to standard genes, often subunits of the ATP synthase. These features are typical for genes associated with CMS in numerous systems (reviewed by Hanson and Bentolila, 2004; Schnable and Wise, 1998). In accordance with these characteristics, orf294 is positioned just upstream of atp1 and orf240a contains a part of the rps3 gene. Interestingly, the predicted amino acid sequence of orf240a also shares some similarities with orf222 and orf224, that has been associated with B. napus nap and pol CMS (Brown, 1999). As suggested by Marienfeld et al. (1997) the proteins encoded by the A. thaliana orfs could be detrimental to mitochondrial function and, as a result, induce CMS. These proteins are normally suppressed by nuclear genes, e.g. by rapid transcript degradation. However, when the mitochondria are transferred to another nuclear background, such as B. napus as described here, the orfs may well be activated. In conclusion, we have shown that several mitochondrial genes/orfs are transcribed in the CMS line, but suppressed when nuclear fragments from the cytoplasmic donor species are present such as, in this case, the A. thaliana chromosome III.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary Material
  9. References
  10. Supporting Information

Plant material

The plant material consists of the B. napus CMS line 4:19 (Leino et al., 2003) back-crossed to B. napus cv. Hanna for nine generations and the restored line 46 which is isogenic with the CMS line with respect to the mitochondrial genome (Leino et al., 2004). Furthermore, the parental species B. napus cv. Hanna and A. thaliana var. Landsberg erecta were included. All plants were grown under controlled conditions in a climate chamber at 22°C day/18°C night temperatures and a photoperiod of 16 h. Light intensity was 400 μmol m−2 sec−1 and air humidity kept at 85%.

Preparation of mitochondria

A crude fraction of mitochondria was isolated by differential centrifugation. Fresh flower buds (3–10 g) were ground in a mortar with 50 ml of grinding buffer containing 0.3 m sucrose, 50 mm MOPS-KOH pH 7.8, 2 mm EDTA, 1% BSA, 1% PVP and 20 mm cysteine. The suspension was filtered through miracloth, centrifuged twice for 3 min at approximately 8 800 g in an SS-34 rotor (Sorvall, Wilmington, DE, USA) and the supernatant was saved. Crude mitochondria were collected by centrifugation for 15 min at approximately 12 000 g in the same rotor. The pellet was resuspended in wash buffer (0.3 m sucrose, 10 mm MOPS-KOH pH 7.2, 1 mm EGTA) by a paintbrush. Run-on experiments and isolation of nucleic acids were performed immediately with the crude mitochondrial suspension.

Run-on transcription

Run-on transcription was performed as described by Giegéet al. (2000) using crude fractions of mitochondria, corresponding to 100 μg protein. After extraction of labelled transcripts with phenol/chloroform (1:1) unincorporated nucleotides were removed by a sepharose CL-6B spin column (AP Biotech, Uppsala, Sweden).

Isolation of nucleic acids

For RNA isolation the crude mitochondrial fraction was lysed in 3 volumes of lysis buffer [250 mm Tris-HCl pH 7.5, 75 mm EDTA, 10% (w/v) sodium N-laurylsarcosine]. Proteins were removed by one phenol/chloroform (1:1) extraction and phenol residues removed by one chloroform extraction. RNAs were precipitated with 1 volume 4 m LiCl overnight at +4°C and then centrifuged at +4°C for 45 min at approximately 9200 g in an SM-24 rotor (Sorvall). The pellet was dissolved in 1 ml water and reprecipitated for 5 h with 1 volume 4 m LiCl at +4°C. After centrifugation for 30 min at approximately 18 000 g in a microcentrifuge the pellet was washed twice in 100% EtOH, resuspended in water and stored at −80°C.

DNA was isolated by lysing the crude mitochondrial fraction with 3 volumes lysis buffer [50 mm Tris-HCl pH 8.0, 20 mm EDTA, 4% (w/v) sodium N-laurylsarcosine] and 0.5 mg ml−1 proteinase K at +60°C for 1 h. Proteins were removed by phenol/chisam extractions and the DNA was RNAse treated and ethanol precipitated.

Kinase labelling of steady-state transcripts and DNA

Labelling of transcripts or DNA fragments at the 5′ ends was performed using T4 polynucleotide kinase (PNK) (MBI Fermentas, Vilnius, Lithuania) in forward reaction. Fifteen micrograms of RNA or 10 μg sonicated (to about 1 kb) DNA was dephosphorylated with calf intestine alkaline phosphatase (MBI Fermentas) in 50 μl of the supplied buffer at 37°C for 30 min. Dephosphorylated transcripts were phenol/chloroform (1:1) extracted twice and ethanol precipitated. Labelling was performed for 30 min at 37°C in a volume of 25 μl in the forward reaction buffer (50 mm Tris-HCl pH 7.6, 10 mm MgCl2, 5 mm DTT, 0.1 mm spermidine and 0.1 mm EDTA), 10 U PNK and 50 μCi [γ-32P]ATP (3000 Ci mm−1). The reaction was stopped by adding 1 μl 0.5 m EDTA pH 8.0. Transcripts/DNA fragments were phenol/chloroform (1:1) extracted once and unincorporated nucleotides were removed by a sepharose CL-6B spin column. As a control run-off transcripts were produced from the pGEM®-T Easy vector (Promega, Madison, WI, USA) restricted with NcoI or XmnI using the in vitro transcription kit MaxiScript from Ambion (Austin, TX, USA).

Filter preparation and hybridization

Mitochondrial DNA sequences, describing gene or exon sequences of all identified protein coding genes in the A. thaliana mt-genome, were PCR amplified from A. thaliana genomic DNA. The primers used were mainly the same as used by Giegéet al. (2000), with the exception for genes nad9, atp6, rps3, rps4, orf262, orf139, orf313, orf215a, orf275a and orf153b. Sequences are found online in Table S1. The PCR products, of an average length of 432 nucleotides, were cloned into the pGEM®-T Easy vector (Promega) and sequenced to verify correct identity. DNA was then PCR amplified once more using the plasmids as template. Amplified DNA from each gene was denatured in 0.2 m NaOH for 15 min at 37°C and spotted twice on Hybond N+ nylon filters (AP Biotech) using a Quadra Model 250 robot (Tomtec, Hamden, CT, USA). Each 1 μl dot contained 250 ng DNA. The membranes were baked at 80°C for 2 h to fix the DNA. To verify the amounts of target molecules available on the filter, a Southern hybridization with the vector arm sequence was performed.

Hybridizations and pre-hybridizations were performed in 0.5 m phosphate buffer pH 7.2, 7% SDS and 10 mm EDTA at 65°C. After RNA-DNA hybridization the membranes were rinsed with 2x SCC, 0.1% SDS and washed at 65°C with 1x SSC, 0.1% SDS for 15 min and twice with 0.2x SSC, 0.1% SDS for 10 min. After washings, run-on assays were treated with 100 U RNAse T1 and 200 μg RNAse A in 10 ml 2x SSC at 37°C for 1 h to remove single-stranded RNA. DNA-DNA hybridization membranes were washed at room temperature twice with 2x SCC, 0.1% SDS for 5 min, twice with 1x SSC, 0.1% SDS for 10 min and twice with 0.1x SSC, 0.1% SDS for 5 min. The membranes were exposed to phosphorimager screens (Molecular Dynamics, Sunnyvale, CA, USA) and scanned on a Personal FX phosphorimager (Bio-Rad, Hercules, CA, USA).

RFLP and Northern analysis

RNA (10 μg) was electrophoresed on 1.5% denaturing agarose gels and transferred to Hybond N+ membranes (AP Biotech) according to the manufacturer's instructions. MtDNA (1 μg) was restricted with EcoRI, BamHI or HindIII, electrophoresed and blotted onto Hybond N+ membranes according to instructions. The probes (the same as used in dot-blots) were labelled with [α-32P]dCTP by random priming using the Random Primed DNA Labeling Kit from Roche (Penzberg, Germany). Hybridization conditions and filter washings were performed as described above.

Quantification and statistical analyses

The signal intensity of each dot was calculated using Quantity One (Bio-Rad) software. In the dot-blot experiments signals from the two dots of each gene were divided with the total signal of all dots in a given experiment. In the run-on assays calculated values were normalized for the amount of UTP or GTP incorporated in each fragment based on the coding strand sequence of the target DNA. Run-on and RNA kinase experiments were repeated four times and DNA kinase experiments three times, with independently labelled RNA or DNA from separate preparations of mitochondria. Statistical analysis of transcriptional activity and transcript abundance variance between lines for individual genes was performed using one-way anova. For genes showing significant F-tests pairwise comparisons were made using Bonferroni-corrected P-values. All calculations were performed in MINITAB®14 software using the general linear model procedure.

Circular RT-PCR

Circular RT-PCR was performed according to the protocol described in Perrin et al. (2004). Five micrograms of mtRNA was incubated with 40 U T4 RNA ligase (NEB) and 1 U RQ1 DNAse (Promega) for 1 h at 37°C. The RNA was phenol/chloroform extracted and ethanol precipitated. cDNA was synthesized from 1 μg circulized RNA using SuperscriptTM III RT (Invitrogen, Carlsbad, CA, USA) and reverse primers (Table S6) for orf240a, orf294 and orf139 respectively. The regions comprising the ligated 5′-3′ ends were then PCR amplified using forward and reverse primers (Table S6) for the respective regions using AmpliTaq Gold (Roche). The amplification protocol consisted of a 10-min initial denaturation at 95°C, 30 cycles of 94°C for 30 sec, 55°C for 30 sec and 72°C for 1 min and a 10-min final elongation step at 72°C. The amplified products were TOPO TA-cloned (Invitrogen) and sequenced. The complete experiment was repeated once from separate RNA preparations.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary Material
  9. References
  10. Supporting Information

The authors thank I. Eriksson and S. Thyselius for excellent technical assistance and Dr P. Giegé for providing primer sequences for amplification of mitochondrial probes. We also thank Prof. A. Brennicke and Dr P. Bergman for valuable comments on the manuscript. This work was supported by the Swedish Research Council (VR), the Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning (FORMAS), the Carl Tryggers Foundation and Martha and Fredrik Nilssons Remembrance Foundation.

Supplementary Material

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary Material
  9. References
  10. Supporting Information

Table S1 Primer sequences used to amplify mitochondrial gene probes used in the investigation

Table S2 Relative transcriptional rates as analysed by run-on experiments. The values are expressed as percentage of the total amount observed in the respective line. Average values ± standard deviation are from four separate experiments

Table S3 Relative stoichiometry of mitochondrial genes in the parental, CMS and restored lines. Values are presented in relation to A. thaliana for each gene. Average values ± standard deviation are from three separate experiments

Table S4 RFLP patterning of the CMS/restored line. The restriction enzyme used is indicated for each gene probe. A, A. thaliana fragment; B, B. napus fragment; R, recombined fragment

Table S5 Relative transcript abundance as analysed by kinase end-labelling experiments. The values are expressed as percentage of the total amount observed in the respective line. Average values ± standard deviations are from four separate experiments

Table S6 Primer sequences for the cRT-PCR experiments

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary Material
  9. References
  10. Supporting Information
  • Abdelnoor, R.V., Yule, R., Elo, A., Christensen, A.C., Meyer-Gauen, G. and Mackenzie, S.A. (2003) Substoichiometric shifting in the plant mitochondrial genome is influenced by a gene homologous to MutS. Proc. Natl Acad. Sci. USA, 100, 59685973.
  • Bellaoui, M., Martin-Canadell, A., Pelletier, G. and Budar, F. (1998) Low-copy-number molecules are produced by recombination, actively maintained and can be amplified in the mitochondrial genome of Brassicaceae: relationship to reversion of the male sterile phenotype in some cybrids. Mol. Gen. Genet. 257, 177185.
  • Binder, S. and Brennicke, A. (2003) Gene expression in plant mitochondria: transcriptional and post-transcriptional control. Philos. Trans. R. Soc. Lond. B Biol. Sci. 358, 181189.
  • Binder, S., Marchfelder, A. and Brennicke, A. (1996) Regulation of gene expression in plant mitochondria. Plant Mol. Biol. 32, 303314.
  • Brennicke, A., Zabaleta, E., Dombrowski, S., Hoffmann, M. and Binder, S. (1999) Transcription signals of mitochondrial and nuclear genes for mitochondrial proteins in dicot plants. J. Hered. 90, 345350.
  • Brown, G. (1999) Unique aspects of cytoplasmic male sterility and fertility restoration in Brassica napus. J. Hered. 90, 351356
  • Dill, C.L., Wise, R.P. and Schnable, P.S. (1997) Rf8 and Rf* mediate unique T-urf13-transcript accumulation, revealing a conserved motif associated with RNA processing in restoration of pollen fertility in T-cytoplasm maize. Genetics, 147, 13671379.
  • Dombrowski, S., Hoffmann, M., Kuhn, J., Brennicke, A. and Binder, S. (1998) On mitochondrial promoters in Arabidopsis thaliana and other flowering plants. In Plant Mitochondria: From Gene to Function (Möller, I.M., Gardeström, P., Glimelius, K. and Glaser, E., eds). Leiden, the Netherlands: Backyuys Publishers, pp. 165170.
  • Earle, E.D. (1995) Mitochondrial DNA in somatic hybrids and cybrids. In The Molecular Biology of Plant Mitochondria (Levings, C.S.III and Vasil, I.K., eds). Dordrecht, the Netherlands: Kluwer Academic Publishers, pp. 557584.
  • Edqvist, J. and Bergman, P. (2002) Nuclear identity specifies transcriptional initiation in plant mitochondria. Plant Mol. Biol. 49, 5968.
  • Fey, J. and Maréchal-Drouard, L. (1999) Compilation and analysis of plant mitochondrial promoter sequences: an illustration of a divergent evolution between monocot and dicot mitochondria. Biochem. Biophys. Res. Commun. 256, 409414.
  • Finnegan, P.M. and Brown, G.G. (1990) Transcriptional and post-transcriptional regulation of RNA levels in maize mitochondria. Plant Cell, 2, 7183.
  • Gagliardi, D. and Leaver, C.J. (1999) Polyadenylation accelerates the degradation of the mitochondrial mRNA associated with cytoplasmic male sterility in sunflower. EMBO J. 18, 37573766.
  • Gagliardi, D., Perrin, R., Maréchal-Drouard, L., Grienenberger, J.-M. and Leaver, C.J. (2001) Plant mitochondrial polyadenylated mRNAs are degraded by a 3′- to 5′-exoribonuclease activity, which proceeds unimpeded by stable secondary structures. J. Biol. Chem. 276, 4354143547.
  • Giegé, P. and Brennicke, A. (2001) From gene to protein in higher plant mitochondria. C. R. Acad. Sci. III, 324, 209217.
  • Giegé, P., Hoffmann, M., Binder, S. and Brennicke, A. (2000) RNA degradation buffers asymmetries of transcription in Arabidopsis mitochondria. EMBO Rep. 1, 164170.
  • Gray, M.W. (1999) Evolution of organellar genomes. Curr. Opin. Genet. Dev. 9, 678687.
  • Håkansson, G. and Glimelius, K. (1991) Extensive nuclear influence on mitochondrial transcription and genome structure in male-fertile and male-sterile alloplasmic Nicotiana materials. Mol. Gen. Genet. 229, 380388.
  • 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. 28, 25712576.
  • Hanson, M.R. and Bentolila, S. (2004) Interactions of mitochondrial and nuclear genes that affect male gametophyte development. Plant Cell, 16, 154169.
  • Janska, H., Sarria, R., Woloszynska, M., Arrieta-Montiel, M. and Mackenzie, S.A. (1998) Stoichiometric shifts in the common bean mitochondrial genome leading to male sterility and spontaneous reversion to fertility. Plant Cell, 10, 11631180.
  • Kuhn, J., Tengler, U. and Binder, S. (2001) Transcript lifetime is balanced between stabilizing stem-loop structures and degradation-promoting polyadenylation in plant mitochondria. Mol. Cell. Biol. 21, 731742.
  • Leino, M., Teixeira, R., Landgren, M. and Glimelius, K. (2003) Brassica napus lines with rearranged Arabidopsis mitochondria display CMS and a range of developmental aberrations. Theor. Appl. Genet. 106, 11561163.
  • Leino, M., Thyselius, S., Landgren, M. and Glimelius, K. (2004) Arabidopsis thaliana chromosome III restores fertility in a cytoplasmic male-sterile Brassica napus line with A. thaliana mitochondrial DNA. Theor. Appl. Genet. 109, 272279.
  • Li, X.Q., Jean, M., Landry, B.S. and Brown, G.G. (1998) Restorer genes for different forms of Brassica cytoplasmic male sterility map to a single nuclear locus that modifies transcripts of several mitochondrial genes. Proc. Natl Acad. Sci. USA, 95, 1003210037.
  • Marienfeld, J.R., Unseld, M., Brandt, P. and Brennicke, A. (1997) Mosaic open reading frames in the Arabidopsis thaliana mitochondrial genome. Biol. Chem. 378, 859862.
  • Marker, C., Zemann, A., Terhörst, T., Kiefmann, M., Kastenmayer, J.P., Green, P., Bachellerie, J.-P., Brosius, J. and Hüttenhofer, A. (2002) Experimental RNomics: identification of 140 candidates for small non-messenger RNAs in the plant Arabidopsis thaliana. Curr. Biol. 12, 20022013.
  • Muise, R.C. and Hauswirth, W.W. (1992) Transcription in maize mitochondria: effects of tissue and mitochondrial genotype. Curr. Genet. 22, 232242.
  • Muise, R.C. and Hauswirth, W.W. (1995) Selective DNA amplification regulates transcript levels in plant mitochondria. Curr. Genet. 28, 113121.
  • Mulligan, R.M., Leon, P. and Walbot, V. (1991) Transcriptional and posttranscriptional regulation of maize mitochondrial gene expression. Mol. Cell. Biol. 11, 533543.
  • Newton, K.J., Winberg, B., Yamato, K., Lupold, S. and Stern, D.B. (1995) Evidence for a novel mitochondrial promoter preceding the cox2 gene of perennial teosintes. EMBO J. 14, 585593.
  • Palmer, J.D. and Herbon, L.A. (1988) Plant mitochondrial DNA evolves rapidly in structure, but slowly in sequence. J. Mol. Evol. 28, 8797.
  • Pelletier, G. (1986) Plant organelle genetics through somatic hybridisation. Oxford Surv. Plant Mol. Cell Biol. 3, 97121.
  • Perrin, R., Lange, H., Grienenberger, J.-M. and Gagliardi, D. (2004) AtmtPNPase is required for multiple aspects of the 18S rRNA metabolism in Arabidopsis thaliana mitochondria. Nucleic Acids Res. 32, 51745182.
  • Remacle, C. and Maréchal-Drouard, L. (1996) Characterization of the potato mitochondrial transcription unit containing ‘native’trnS (GCU), trnF (GAA) and trnP (UGG). Plant Mol. Biol. 30, 553563.
  • Schnable, P.S. and Wise, R.P. (1998) The molecular basis of cytoplasmic male sterility and fertility restoration. Trends Plant Sci. 3, 175180.
  • Singh, M., Hamel, N., Menassa, R., Li, X.Q., Young, B., Jean, M., Landry, B.S. and Brown, G.G. (1996) Nuclear genes associated with a single Brassica CMS restorer locus influence transcripts of three different mitochondrial gene regions. Genetics, 143, 505516.
  • Unseld, M., Marienfeld, J.R., Brandt, P. and Brennicke, A. (1997) The mitochondrial genome of Arabidopsis thaliana contains 57 genes in 366 924 nucleotides. Nat. Genet. 15, 5761.
  • Wen, L. and Chase, C.D. (1999) Pleiotropic effects of a nuclear restorer-of-fertility locus on mitochondrial transcripts in male-fertile and S male-sterile maize. Curr. Genet. 35, 521526.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. Supplementary Material
  9. References
  10. Supporting Information

Table S1.  Primer sequences used to amplify mitochondrial gene probes used in the investigation.

Table S2.  Relative transcriptional rates as analysed by run-on experiments. The values are expressed as percentage of the total amount observed in the respective line. Average values ± standard deviation are from four separate experiments.

Table S3.  Relative stoichiometry of mitochondrial genes in the parental, CMS and restored lines. Values are presented in relation to A. thaliana for each gene. Average values ± standard deviation are from three separate experiments.

Table S4.  RFLP patterning of the CMS/restored line. The restriction enzyme used is indicated for each gene probe. A, A. thaliana fragment; B, B. napus fragment; R, recombined fragment.

Table S5.  in a given experiment. In the run-on assays calculated values were normalized for the amount of UTP or GTP incorporated in each fragment based on the coding strand sequence of the target DNA. Relative transcript abundance as analysed by kinase endlabelling experiments. The values are expressed as percentage of the total amount observed in the respective line. Average values ± standard deviations are from four separate experiments.

Table S6.  Primer sequences for the cRT-PCR experiments.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
TPJ_2389_sm_suppledata.doc340KSupporting info item
TPJ_2389_sm_TableS1.doc30KSupporting info item
TPJ_2389_sm_TableS2.doc168KSupporting info item
TPJ_2389_sm_TableS3.doc133KSupporting info item
TPJ_2389_sm_TableS4.doc46KSupporting info item
TPJ_2389_sm_TableS5.doc178KSupporting info item
TPJ_2389_sm_TableS6.doc24KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.