Rice MPR25 encodes a pentatricopeptide repeat protein and is essential for RNA editing of nad5 transcripts in mitochondria


  • Takushi Toda,

    1. Laboratory of Environmental Plant Biotechnology, Graduate School of Agricultural Science, Tohoku University, 1-1 Tsutsumidori-Amamiyamachi, Aoba-ku, Sendai 981-8555, Japan
    Search for more papers by this author
  • Sota Fujii,

    1. Laboratory of Environmental Plant Biotechnology, Graduate School of Agricultural Science, Tohoku University, 1-1 Tsutsumidori-Amamiyamachi, Aoba-ku, Sendai 981-8555, Japan
    Search for more papers by this author
    • Present address: Graduate School of Science, Kyoto University, Oiwakecho, Kitashirakawa, Sakyo-ku, Kyoto 606-8502, Japan.

  • Ko Noguchi,

    1. Department of Biological Sciences, Graduate School of Science, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
    Search for more papers by this author
  • Tomohiko Kazama,

    1. Laboratory of Environmental Plant Biotechnology, Graduate School of Agricultural Science, Tohoku University, 1-1 Tsutsumidori-Amamiyamachi, Aoba-ku, Sendai 981-8555, Japan
    Search for more papers by this author
  • Kinya Toriyama

    Corresponding author
    1. Laboratory of Environmental Plant Biotechnology, Graduate School of Agricultural Science, Tohoku University, 1-1 Tsutsumidori-Amamiyamachi, Aoba-ku, Sendai 981-8555, Japan
      (e-mail torikin@bios.tohoku.ac.jp).
    Search for more papers by this author

(e-mail torikin@bios.tohoku.ac.jp).


Pentatricopeptide repeat (PPR) proteins are involved in the modification of organelle transcripts. In this study, we investigated the molecular function in rice of the mitochondrial PPR-encoding gene MITOCHONDRIAL PPR25 (MPR25), which belongs to the E subgroup of the PPR family. A Tos17 knockout mutant of MPR25 exhibited growth retardation and pale-green leaves with reduced chlorophyll content during the early stages of plant development. The photosynthetic rate in the mpr25 mutant was significantly decreased, especially under strong light conditions, although the respiration rate did not differ from that of wild-type plants. MPR25 was preferentially expressed in leaves. FLAG-tagged MPR25 accumulated in mitochondria but not in chloroplasts. Direct sequencing revealed that the mpr25 mutant fails to edit a C–U RNA editing site at nucleotide 1580 of nad5, which encodes a subunit of complex I (NADH dehydrogenase) of the respiratory chain in mitochondria. RNA editing of this site is responsible for a change in amino acid from serine to leucine. Recombinant MPR25 directly interacted with the proximal region of the editing site of nad5 transcripts. However, the NADH dehydrogenase activity of complex I was not affected in the mutant. By contrast, genes encoding alternative NADH dehydrogenases and alternative oxidase were up-regulated. The mpr25 mutant may therefore provide new information on the coordinated interaction between mitochondria and chloroplasts.


Pentatricopeptide repeat (PPR) proteins are characterized by a tandem array of a degenerate 35 amino acid repeat, and form a huge gene family in land plants. There are 450 members in Arabidopsis and 477 members in rice (Small and Peeters, 2000; Andres et al., 2007; Saha et al., 2007; O’Toole et al., 2008). It has been predicted that half of PPRs are targeted to mitochondria, and one-quarter localize to chloroplasts (Lurin et al., 2004). PPR proteins are often essential for post-transcriptional RNA regulation processes in mitochondria or chloroplasts, such as RNA editing (Hammani et al., 2009; Kim et al., 2009; Tang et al., 2009; Zehrmann et al., 2009; Fujii and Small, 2011), RNA processing (Fisk et al., 1999; Kazama and Toriyama, 2003; Meierhoff et al., 2003; Nakamura et al., 2003; Nakamura and Sugita, 2008), RNA splicing (Schmitz-Linneweber et al., 2006; Falcon de Longevialle et al., 2007, 2008; Hattori et al., 2007), RNA stabilization (Beick et al., 2008; Tavares-Carreon et al., 2008) and translational activation (Williams and Barkan, 2003; Schmitz-Linneweber et al., 2005; Pusnik et al., 2007; Uyttewaal et al., 2008; Davies et al., 2009; Rackham et al., 2009). As PPRs are large proteins that are highly degenerate, it is presumed that each PPR has its own specific RNA target (Prikryl et al., 2011). PPR mutants exhibit various phenotypes, such as developmental defects (Falcon de Longevialle et al., 2007, 2008), reduced fertility (Lurin et al., 2004) and embryo lethality (Gutierrez-Marcos et al., 2007). There are only a few reports of PPR proteins from rice compared with those from Arabidopsis (Kazama and Toriyama, 2003; Gothandam et al., 2005; Kazama et al., 2008; Kim et al., 2009; Su et al., 2012). To reveal the function of mitochondrial PPR proteins in rice, we previously performed screening of rice containing mutated mitochondrial PPR genes from the Tos17 insertion mutant line database (http://tos.nias.affrc.go.jp/), and found 16 PPR genes with high probabilities of Tos17 insertion (Toda et al., 2010).

In the present study, we focused on one of the Tos17 knockout mutants, in which a gene in locus LOC_Os04g51350, named MITOCHONDRIAL PPR25 (MPR25), was disrupted by a retrotransposon insertion. We show that MPR25 is required for C–U RNA editing of nad5 transcripts in mitochondria. The plant developmental phenotype, photosynthetic and respiratory activity, and expression of genes for alternative respiratory pathways were also characterized.


The mpr25 mutant exhibits a mild phenotype of pale-green leaves and growth retardation

MPR25 encodes an 805 amino acid protein with 16 PPR motifs and an E domain at its C-terminus, based on Pfam motif analysis (http://pfam.sanger.ac.uk./search?tab=searchSequenceBlock) (Figure 1a). These 16 PPR motifs are grouped into five canonical PPR motifs (P motif), six PPR-like S motifs (short) and five PPR-like L motifs (long) (Lurin et al., 2004). In the Tos17 insertion line NC0057, Tos17 was inserted at nucleotide 1327 from the initiation codon in the ninth PPR motif of MPR25 (Figure 1a). Quantitative RT-PCR analysis of MPR25 mRNA showed that MPR25 was highly expressed in green leaves of a wild-type (WT) plant (Figure 1b) and weakly so in leaves of etiolated seedlings, roots, pollen, pistils, seeds and calli. In contrast, the amount of mRNA in the green leaves of the mutant was <5% of that in the WT (Figure 1c). Due to the disruption in the ninth PPR motif and reduced mRNA levels, we considered the mutant allele to be a null allele. The segregation of +/− (hemizygously carrying the Tos17 inserted allele) self-fertilizing progeny was 129 +/+: 217 +/−: 99 −/− (equivalent to 1:2:1, χ2 = 4.32, 0.1<P<0.2), indicating that no haploid lethality or seed abortion phenotypes were caused by the mpr25 mutation. The homozygous mpr25 mutant exhibited growth retardation (Figure 1d). The maximum shoot length of 1-week-old seedlings of the homozygous mpr25 plants under natural light conditions was indistinguishable from that of the WT; however, shoots of 2–11-week-old seedlings homozygous for mpr25 were significantly shorter than those of WT (Figure 2a). Growth retardation was also observed when the plants were grown in the dark (Figure S1). The maximum shoot length of the mpr25 mutant was significantly shorter than that of WT in 2- and 3-week-old etiolated seedlings grown in the dark (Figure 2b).

Figure 1.

 MPR25 protein structure, gene expression and phenotype of the mpr25 mutant.
(a) Predicted protein structure of MPR25, and the position of the Tos17 insertion. Blue boxes indicate S motifs, orange boxes indicate P motifs, gray boxes indicate L motifs, and the green box indicates the E domain.
(b) Quantitative RT-PCR analysis of MPR25 expression in green leaves and roots from 10-week-old seedlings, etiolated leaves of 4-week-old seedlings, pollen, pistils, seeds and calli. Transcript abundance is indicated relative to that of calli.
(c) Quantitative RT-PCR analysis of MPR25 expression in green leaves of 4-week-old seedlings of WT and the mpr25 mutant. Transcript abundance is indicated relative to that of WT. The asterisk indicates a statistically significant difference between WT and the mpr25 mutant at < 0.05 (Student’s t-test).
(d) Phenotypes of the mpr25 mutant. Plants were grown on soil for 4 weeks. Scale bar = 6 cm.
(e) Complemented mpr25 mutant transformed with Ubi-MPR25. The mpr25 plant was transformed with Ubi-GUS as a vector control. The photographs are of 4-week-old seedlings of the T1 generation. Scale bar = 10 cm.

Figure 2.

 Growth analysis of the mpr25 mutant.
(a.b) Maximum shoot length of WT and the mpr25 mutant under (a) natural light conditions (n = 4) and (b) dark conditions (n = 10).
(c) SPAD values for the WT and mpr25 mutant under natural light conditions (n = 4).
(d) Number of tillers of the WT and mpr25 mutant (n = 4).
Asterisks indicate statistically significant differences between WT and the mpr25 mutant at < 0.05 (Student’s t-test).

Under natural light conditions, the mutant exhibited pale-green leaves (Figure 1d). The decreased intensity of green color was confirmed by the reduction of chlorophyll content (SPAD value), which was determined using a chlorophyll meter. The SPAD values of the mutant were significantly lower than those of WT in 2–5-week-old seedlings (Figure 2c). These results indicate that the mpr25 mutant is deficient in chlorophyll content.

Although the mpr25 mutant showed shorter shoot length and lower chlorophyll content until 11 weeks after germination, these deficiencies recovered to the normal state after 13 weeks (Figure 2a,c). After the flowering stage at 16 weeks, the maximum shoot length and SPAD value were almost identical in the mpr25 mutant and WT. The maximum shoot length was 97.3 ± 0.7 cm in the mpr25 mutant and 101.3 ± 0.7 cm in the WT at 16 weeks. The SPAD value was 41.7 ± 0.2 in the WT and 34.8 ± 2.1 in the mpr25 mutant. Although these parameters recovered in the mutant at the flowering stages, the number of tillers remained suppressed (Figure 2d), with 1–3 being present in the mpr25 mutant compared with 12 in the WT at 16 weeks after germination (Figure 2d). Reduced tiller branching was probably caused by the initial growth retardation. The flowering stage was retarded by 2 weeks in the mpr25 mutant.

A 2415 bp DNA fragment corresponding to the full-length MPR25 open reading frame (ORF) under the control of the ubiquitin promoter (Ubi-MPR25) was introduced into the mpr25 mutant for the complementation test. All 10 transgenic plants expressing Ubi-MPR25 recovered normal development, in contrast to six transgenic plants expressing Ubi-GUS, which did not show any complementation of the mutant phenotype (Figure 1d). The complemented phenotype was stably inherited in the T1 plants. In the T1 generation, the maximum shoot length of 8-week-old plants was 24.1 ± 3.9 cm in the mpr25 plants transformed with Ubi-GUS and 44.7 ± 0.7 cm in the transgenic plants complemented with Ubi-MPR25, compared with 45.1 ± 1.0 cm in WT. The SPAD value was 38.0 ± 1.5 in WT, 22.4 ± 0.5 in the mpr25 plants expressing Ubi-GUS, and 37.3 ± 0.7 in the complemented mpr25 plants expressing Ubi-MPR25. These results confirm that the mutant phenotype of reduced shoot length and low chlorophyll content is caused by disruption of MPR25.

The photosynthesis rate is lower in mpr25 mutants

To investigate whether photosynthesis was affected in the mpr25 mutant, we first measured the CO2 assimilation rate at a light intensity of 80, 300 or 1200 μmol photon m−2 sec−1). The CO2 assimilation rate was lower in the mpr25 mutant than in WT (Figure 3a). The difference was particularly significant at light intensities of 300 and 1200 μmol m−2 sec−1. However, the stomatal conductance was not significantly different (Figure 3b). These results indicate that the photosynthesis system was affected in the mpr25 mutants. We then examined photosynthesis electron transport using measurements of chlorophyll fluorescence (Baker, 2008) and P700 redox state (Yoshida et al., 2007). Fv/Fm, the maximum operating efficiency of photosystem II (PSII), was not affected in the mpr25 mutant (Figure S2), but the quantum yield of PSII (ΦII) was decreased in the mutant (Figure S2). This decrease was caused by the increased value for non-photochemical quenching and the decreased value for photochemical quenching (Figure S2). These results indicate that the heat dissipation efficiency was up-regulated by a process in photosynthesis electron transport downstream of PSII that is impaired in the mpr25 mutant. The decrease of ΦII in the mutant led to increase in the P700 oxidation ratio (Figure S2). By contrast, the quantum yield of PSI (ΦI) in the mutant was similar to that in WT (Figure S2).

Figure 3.

 Photosynthesis activity of WT and the mpr25 mutant.
(a) CO2 assimilation rate under various light intensities.
(b) Stomatal conductance.
Samples are leaves from 4-week-old seedlings grown under 50% shade (n = 4). Asterisks indicate statistically significant differences between WT and the mpr25 mutant at < 0.05 (Student’s t-test).

MPR25 protein is targeted to mitochondria

Green fluorescent protein (GFP) was fused to the C-terminus of MPR25, and the fusion protein was transiently expressed in epidermal cells of Tulipa gesneriana leaves by particle bombardment assay (Figure 4a). The MPR25–GFP fusion protein expression construct was co-bombarded with a construct expressing red fluorescent protein (RFP) protein fused at the C-terminus of the targeting signal of the F1F0 ATPase γ-subunit. The fluorescence of GFP overlapped with that of RFP, but did not co-localize with the auto-fluorescence of chloroplasts. (Figure 4a). To further confirm the subcellular localization of MPR25, MPR25 was fused to FLAG and stably expressed in transgenic rice cells. Immunoblot analysis of transgenic calli and leaves using anti-FLAG antibody detected the MPR25–FLAG fusion protein in the mitochondrial fraction extracted from calli, where we detected isocitrate dehydrogenase (IDH), which is known to localize in the mitochondrial matrix. MPR25–FLAG was not detected in the chloroplast fraction isolated from leaves, suggesting that MPR25 accumulates exclusively in mitochondria.

Figure 4.

 Subcellular localization of MPR25.
(a) The constructs CaMV 35S promoter-MPR25-GFP and CaMV 35S promoter-F1F0 ATPase targeting signal-RFP were transiently expressed in epidermal cells of Tulipa gesneriana leaves. Green fluorescence represents GFP, red fluorescence represents RFP, yellow fluorescence represents merged images of GFP and RFP, and blue fluorescence represents chloroplast fluorescence. The panel labeled ‘All signals’ shows merged images of GFP, RFP and chloroplast fluorescence.
(b) Immunoblot detection of MPR25-FLAG in transgenic rice. Total mitochondrial (Mt) and chloroplast (Cp) proteins (4 mg each) were loaded. The large subunit of ribulose-1,5-bisphosphate carboxylase/oxygenase (RbcL) and isocitrate dehydrogenase (IDH) were detected as controls for the chloroplast and mitochondrial fractions, respectively.

MPR25 is involved in C–U RNA editing of nad5 transcripts

PPR proteins bind to RNA in a sequence-specific manner and are involved in post-transcriptional RNA splicing, RNA editing or translational regulation. Most of the PPR proteins in the E, E/E+ and DYW subgroups are known to be involved in C–U RNA editing (Fujii and Small, 2011). MPR25 belongs to the E subgroup, and thus is expected to be involved in RNA editing. In young rice seedlings, 491 sites have been reported to be edited in mitochondrial RNA and 21 sites in chloroplast RNA (Corneille et al., 2000; Notsu et al., 2002). Of these, 472 editing sites are located within 34 protein-coding regions in mitochondria, while 20 sites within 10 protein-coding regions and one site within an untranslated region in chloroplasts. We performed direct sequence analysis of all 472 sites within 34 coding regions of mitochondrial cDNA and all 21 sites of chloroplast cDNA, using etiolated leaves of 4-week-old seedlings as a source of RNA. The primers used for RT-PCR and sequence analysis are listed in Table S1. cDNA sequence comparison between the mpr25 mutant and WT revealed that one site within the nad5 transcripts was not edited in the mutant (Figure 5). Nucleotide 1580 in nad5 cDNA was edited to T in WT, but was left as C in the mpr25 mutant. An editing event at this site causes an amino acid change from serine to leucine. Based on the RNA editing site nomenclature proposed by Rüdinger et al. (2009), we named this site ‘nad5eU1580SL’. The absence of an editing event at nad5eU1580SL was also observed in green leaves from the mpr25 mutant (Figure 5). No changes in RNA editing efficiency were found at the other 471 sites in mitochondria, or the 21 sites in chloroplasts from etiolated leaves or green leaves. RNA editing at nad5eU1580SL was recovered in mpr25 transgenic plants expressing Ubi-MPR25, confirming that RNA editing of this site is dependent on MPR25 function.

Figure 5.

 Identification of unedited C residue in the mpr25 mutant, and amino acid sequences of NAD5 in various plant species.
(a) Sequencing chromatograms derived from direct sequencing of RT-PCR products around nucleotide 1580 of nad5.
(b) Sequence alignment of amino acids around nucleotide 1580 of nad5 (BA000029.3 for Oryza sativa, NC_001284 for Arabidopsis thaliana, AP006444 for Brassica napus, NC_007945 for Physcomitrella patens and NC_001638 for Chlamydomonas reinhardtii).

nad5eU1580SL editing is conserved in various plant species

As mentioned above, C–U RNA editing at nad5eU1580SL causes an amino acid replacement of serine by leucine. To estimate the importance of amino acid replacement for NAD5 function, we compared amino acid sequences of NAD5 from Oryza sativa, Arabidopsis thaliana, Brassica napus, Physcomitrella patens and Chlamydomonas reinhardtii. The amino acid sequences deduced from cDNA were conserved, with leucine present at the nad5eU1580SL site in each of the plant species examined (Figure 5b). C–U RNA editing of nad5 RNA occurs in A. thaliana and B. napus at the same sites as O. sativa nad5eU1580SL (Giege and Brennicke,1999; Handa, 2003), whereas the genomic sequences encode leucine in P. patens and C. reinhardtii (Figure 5b). These results indicate that nad5eU1580SL RNA editing and encoding of leucine at this site is probably important for NAD5 function.

MPR25 was expressed at an extremely high level in green leaves compared with other tissues and etiolated leaves (Figure 1b). To investigate whether nad5 was also highly transcribed in green leaves, and whether C–U RNA editing at the nad5eU1580SL site was specific to green leaves, we performed quantitative RT-PCR and direct sequence analysis of nad5 RNA in various WT tissues. Similar amounts of nad5 mRNA and complete C–U RNA editing at the nad5eU1580SL site were detected in leaves and roots of 4-week-old green seedlings, anthers, calli, seeds and leaves of etiolated seedlings (Figure S3). Thus, nad5 was not exclusively expressed in green leaves, and the C–U RNA editing event was not restricted to green leaves, indicating that low expression of MPR25 is sufficient to accomplish C–U RNA editing of nad5 in all tissues.

We also investigated whether the steady-state amounts of nad5 mRNA were affected by disruption of MPR25. Quantitative RT-PCR showed that the amount of nad5 mRNA was not significantly different between WT and the mpr25 mutant in leaves of etiolated or green seedlings (Figure S4). Northern blot analysis in the etiolated seedlings showed that nad5 was expressed at a slightly higher level in the mpr25 mutant than in WT (Figure S4). These results indicate that nad5 mRNA stability was not significantly affected by the absence of C–U RNA editing at nad5eU1580SL in the mpr25 mutant.

Recombinant MPR25 proteins bind to nad5 transcripts

To confirm that the cis sequence of the nad5eU1580SL editing site is the direct target of the MPR25 protein, we performed electrophoresis mobility shift assays (EMSA). Mature MPR25 protein without N-terminal mitochondrial-targeting peptides was expressed in Escherichia coli in fusion with thioredoxin (Trx). The Trx-fused recombinant MPR25 protein (Trx-rMPR25) was incubated with digoxigenin-labeled RNA nucleotides spanning 35 nucleotide regions of nad5eU1580SL. The RNA probe included a sequence from −26 bp upstream to +8 bp downstream of the nad5eU1580SL editing site, excluding any other editing sites. Retardation of the RNA signal was observed only when the nad5 RNA probe was incubated with Trx-rMPR25, but not with Trx alone (Figure 6). We next performed competitive EMSA using non-labeled RNA probe with the same sequence. The signal intensity of the band decreased as the competitor concentration increased, suggesting specific interaction of MPR25 with the probe sequence. The shifted band almost disappeared when a 100-fold excess of the competitor was added (Figure 6). In contrast, retardation of the RNA signal was not observed when the nad7 RNA probe was incubated with Trx-rMPR25 (Figure 6). These results indicate that MPR25 is a trans-acting factor of nad5eU1580SL that directly interacts with the proximal region of the editing site of nad5 transcripts.

Figure 6.

 A recombinant MPR25 protein binds to the 35 nucleotide nad5 RNA probe spanning the nad5eU1580SL editing site.
Trx-rMPR25 is a recombinant protein expressed as a fusion protein with thioredoxin (Trx), and 100 fmol recombinant protein was loaded per lane. The RNA probe was labeled with digoxigenin, and non-labeled RNA probe was used as a competitor. The RNA binding probe and non-binding probe are indicated as ‘shifted’ and ‘free’, respectively.

NADH dehydrogenase activity of complex I was not affected

NAD5 is a subunit of NADH dehydrogenase complex I in the mitochondrial respiratory chain. To investigate whether the respiration rate was affected by the amino acid change in NAD5 due to the absence of RNA editing at nad5eU1580SL, we measured the O2 consumption rate of green leaves sampled from 4-week-old seedlings. The mean respiration rate of the mpr25 mutant was not significantly different from that of WT (Figure 7a).

Figure 7.

 Respiration rate and NADH dehydrogenase activity of complex I.
(a) Respiration rate of leaves of 4-week-old seedlings (n = 4).
(b) NADH dehydrogenase activity of complex I. Mitochondrial membrane complexes isolated from calli were separated on a blue native gel. The gel was stained with Coomassie blue (left). NADH dehydrogenase activity was detected using NADH as a substrate (right). One milligram of protein was loaded per lane (×1). Diluted samples (×1/2 and ×1/5) were loaded for comparison of band intensities. I, II, III, IV and V indicate the position of the respective respiratory chain complexes.

We then examined the activity of NADH dehydrogenase of complex I by a histochemical reaction using NADH as a substrate, after separation of mitochondrial respiratory complexes from calli by blue native PAGE. The result indicates that the activity of NADH dehydrogenase did not differ between WT and the mpr25 mutant (Figure 7b).

Internal/external NADH dehydrogenase and AOX genes were up-regulated

Plant mitochondria are known to have alternative dehydrogenases, namely internal and external NADH dehydrogenases (Rasmusson and Wallstrom, 2010). We performed an expression analysis of the internal NADH dehydrogenase genes NDB1, NDB2 and NDB3 and the external NADH dehydrogenase genes NDA1, NDA2 and NDC1 by quantitative RT-PCR. NDA1, NDB1, NDB2, NDB3 and NDC1 were expressed at a higher level in the leaves of 4-week-old seedlings of the mpr25 mutant compared with WT (Figure S5).

We also analyzed the expression of ALTERNATIVE OXIDASE (AOX) genes, which are known to be mitochondrial stress markers (Gutierres et al., 1997; Sabar et al., 2000). Quantitative RT-PCR analysis showed that AOX1a transcripts in the leaves of 4-week-old seedlings of the mpr25 mutant increased fivefold in comparison with the level in WT, and that the level of AOX1c was doubled in comparison with the level in WT (Figure S6).


MPR25 is involved in nad5eU1580SL editing on mitochondrial transcripts

Mitochondrial respiratory chain complex I in plants is a multimeric enzyme of more than 40 subunits, encoded by both nuclear and mitochondrial genes (Klodman et al., 2010). Rice mitochondrial DNA encodes nine subunits of complex I (NAD1, NAD2, NAD3, NAD4, NAD4L, NAD5, NAD6, NAD7 and NAD9). The rice nad5 transcript comprises a protein-coding region of 2010 nucleotides, and is known to contain 11 editing sites at the nucleotides 1490, 1550, 1580, 1589, 1859, 1895, 1900, 1901, 1916, 1918 and 1958. MPR25 was shown to be involved in C–U RNA editing at nucleotide 1580 (nad5eU1580SL) (Figures 5 and S3), but disruption of MPR25 did not affect the editing efficiency of other sites in nad5. There have been a few reports on editing factors of nad5 transcripts: A. thaliana MEF8 and MEF29, Physcomitrella patens PpPPR_91 and PpPPR_79, and maize PPR2263 (Ohtani et al., 2010; Takenaka et al., 2010; Uchida et al., 2011; Sosso et al., 2012). MEF8 has been reported to be involved in editing nucleotide 676 of Arabidopsis nad5. Mutants with MEF8 disruption did not exhibit any phenotypic changes (Takenaka et al., 2010). Physcomitrella PpPPR_91 was reported to be the editing factor for nucleotide 598 (C) within the nad5 transcript, whereas PpPPR_79 was reported to be involved in editing of nucleotide 730 (C) in the nad5 transcript. Ppppr_91 and Ppppr_79 mutants exhibited severe growth retardation (Ohtani et al., 2010; Uchida et al., 2011). Maize PPR2263 and Arabidopsis MEF29 encode DYW domain-containing PPR proteins, and are required for RNA editing of nad5 and cob transcripts at nucleotides 1550 and 908, respectively (Sosso et al., 2012). The ppr2263 mutation was reported to cause growth defects in kernels and seedlings of maize. The rice nad5eU1580SL site is also known to be edited in A. thaliana (Figure 5). The Arabidopsis MPR25 ortholog is At3g22150, although the function of this gene in Arabidopsis has not yet been reported.

Reports on rice PPR mutants have been limited to OsPPR1 (Gothandam et al., 2005) and OGR1 (Kim et al., 2009). OsPPR1 was reported to be related to chloroplast development, but its molecular function remains to be elucidated (Gothandam et al., 2005). The ogr1 mutant shows growth retardation and fewer tillering. OGR1 belongs to the DYW subgroup, and was shown to be localized to mitochondria and involved in C–U RNA editing at multiple sites in cox2, cox3, ccmC, nad2 and nad4 transcripts (Kim et al., 2009). The mpr25 mutant and the ogr1 mutant showed growth retardation in that the defect of C–U RNA editing of the mitochondrial gene is considered to be involved in plant development.

MPR25 is localized to mitochondria but also related to chloroplast function

RNA editing of the nad5eU1580SL site in mitochondria was abolished in the mpr25 mutant, resulting in lack of the serine to leucine amino acid substitution (Figure 5). This leucine residue is conserved in various plant species and is probably important for NAD5 function (Figure 5). However, the NADH dehydrogenase activity of complex I and the respiration rate of the mpr25 mutant were not significantly different from those of WT (Figure 7). By contrast, genes for alternative NADH dehydrogenases and alternative oxidase were up-regulated in the mutants (Figures S5 and S6). Such alternative pathways, which bypass proton transport across the membrane, may affect ATP production in the electron transport chain. Defects in the mitochondrial electron transport chain are likely to be responsible for the growth retardation phenotype of the mpr25 mutant. Similar defects in plant growth have also been reported in other complex I mutants (Gutierres et al., 1997; Brangeon et al., 2000; Sabar et al., 2000; Lee et al., 2002; Pineau et al., 2005; Falcon de Longevialle et al., 2007; Kim et al., 2009; Sung et al., 2010). For example, a nad1 intron 1 splicing-deficient mutant, otp43, of Arabidopsis exhibited delayed development and flowering phenotype (Falcon de Longevialle et al., 2007); the slo1 mutant of Arabidopsis also showed slow growth and late germination. SLO1 encodes an E group PPR protein, and requires C–U RNA editing at nucleotide 449 of nad4 and nucleotide 328 of nad9 (Sung et al., 2010). MEF9, an E subclass PPR protein, is required for an RNA editing event in nad7 transcripts in Arabidopsis. Mutant mef9-2 displayed slow growth, with bolting and flower set delayed by 2 weeks (Takenaka, 2010). Growth retardation was also reported in the slg mutant of Arabidopsis. SLG1 belongs to the E subgroup of PPRs, and was shown to participate in C–U RNA editing of nad3 transcripts (Yuan and Liu, 2012). However, a defective phenotype was not apparent in the case of a mef1 mutant, although MEF1 encodes a DYW group PPR protein and C–U RNA editing was abolished at editing sites in nad2, nad7 and rps4 transcripts of the mef1 mutant (Zehrmann et al., 2009). Complementation of the respiratory rate by an increase in the activity of alternative NADH dehydrogenases and AOX capacity has also been reported in the Nicotiana sylvestris cytoplasmic male sterile (CMS) II mutant, which lacks nad7 (Pineau et al., 2005), and the nadfs4 mutant, which lacks complex I (Meyer et al., 2009).

The photosynthetic rate was significantly reduced in the mpr25 mutants (Figure 3). Similar reduction of photosynthetic activity has also been reported in a CMSII mutant lacking nad7 (Pineau et al., 2005), and in an NMS1 mutant, which lacks splicing of the first intron of nad4 (Sabar et al., 2000). However, the reasons why photosynthesis is affected by defects in mitochondrial complex I are still unclear. In the nadfs4 mutant of Arabidopsis lacking complex I, the activity of PSII and the electron transport rate were significantly reduced, especially under low irradiance, with non-photochemical quenching being enhanced (Meyer et al., 2009). The photosynthesis rate was more severely impaired when non-photochemical quenching was enhanced under increased light conditions in the mpr25 mutant. NAD5 regulation by MPR25 may be considered to be involved in processes related to photorespiration or the malate/oxaloacetic acid shuttle, which has a role in the dissipation of excess reducing power from chloroplasts (Yoshida et al., 2007; Noguchi and Yoshida, 2008). Alternatively, defects of mitochondrial complex I may have some retrograde action that influences nuclear gene expression of some plastid functions; in addition, impaired metabolite shuttling from the chloroplasts to the mitochondria may induce a stress on photosynthetic activity in mutant plants (Rasmusson and Wallstrom, 2010). Further physiological and biochemical analysis of the mpr25 mutant will elucidate new aspects of the coordinated interaction between mitochondria and chloroplasts.

Experimental Procedures

Genotyping and isolation of a Tos17-inserted line

The Tos17-inserted line, NC0057, of O. sativa L. cv. Nipponbare was provided by Dr Akio Miyao and Dr Hirohiko Hirochika (National Institute of Agrobiological Science, Tsukuba, Japan). The allele in which Tos17 was inserted was PCR-amplified using the MPR25 original forward primer 5′-GGAGTTGAAAAGAGCAAGCG-3′ and Tos17-specific primer 5′-ATTGTTAGGTTGCAAGTTAGTTAAGA-3′. WT alleles were amplified using MPR25 original forward primer and MPR25 original reverse primer 5′-GATCGGTAAACAAGCATG-3′.

Complementation test

For the complementation test, the full-length MPR25 ORF was PCR-cloned into pGEM T-vector (Promega, http://www.promega.com/) using primers 5′-AGATCTTGAGGAGAGATGACTCGGATGA-3′ (BglII site underlined) and 5′-AGATCTTCTATCCGTACTGTCGGCATC-3′. The PCR products were digested with BglII and inserted downstream of the maize ubiquitin promoter in a DHA6His binary vector (Kagaya et al., 2002). The resulting vector, Ubi-MPR25, was introduced into the Tos17-inserted mpr25 mutant and WT cultivar Nipponbare using Agrobacterium-mediated transformation (Toki et al., 2006). Ubi-GUS was used as a vector control.

Quantitative RT-PCR

Total RNAs were purified from the leaves and roots of 4-week-old seedlings, and from pollen, stigmas, seeds, calli and leaves of 4-week-old etiolated seedlings using an RNeasy plant mini kit (Qiagen, http://www.qiagen.com/) according to the manufacturer’s instructions. RNA (10 μg) was treated with DNase I (TaKaRa Bio, http://www.takarabio.com/) and the resulting RNA was reverse-transcribed using ReverTra Ace® (Toyobo, http://www.toyobo-global.com/) using a Not I-dT primer or random primer (TaKaRa Bio). Quantitative RT-PCR was performed using SYBR Green II (TaKaRa Bio) and a Thermal Cycler Dice® Real Time System TP800 (TaKaRa Bio). Primer pairs used for the expression analysis are listed in Table S1.

Measurement of maximum shoot length, SPAD value and number of tillers

Plants were grown in soil in a pot (16.8 cm diameter × 19.5 cm depth) under natural light conditions. Five-week-old plants were transplanted to soil and were grown in 50% shade. The maximum shoot length was measured once a week (n = 4). The chlorophyll content of the fully expanded leaf blades was measured using a SPAD chlorophyll meter (SPAD-502, Konica Minolta, http://www.konicaminolta.com/) (n = 9) at five spots per plant. The SPAD value was calculated based on the difference between two optical densities at wavelengths of 600–700 nm and >700 nm.

Measurement of CO2 assimilation rate, chlorophyll fluorescence, P700 redox state and respiration rate

The CO2 assimilation rate, chlorophyll fluorescence and P700 redox state were measured as previously described (Yoshida et al., 2007), using a pulse-amplitude modulation fluorometer (PAM-101, Walz, http://www.walz.com/) and ED-P700WD-E (Walz). Leaf respiration rates were measured using a Clark-type O2 electrode (Rank Brothers, http://www.rankbrothers.co.uk/), as previously described (Yoshida et al., 2007). The detached leaves were kept above the stirrer bar and electrode surface by a piece of nylon mesh. The reaction medium contained 50 mm HEPES, 10 mm MES (pH 6.6) and 0.2 mm CaCl2. Samples were leaves from 4-week-old seedlings grown under 50% shade (n = 4).

Subcellular localization

To study subcellular localization, the full-length MPR25 ORF was PCR-amplified from genomic DNA of Nipponbare using primers 5′-GGATCCTGAGGAGATGACTCGGATGA-3′ (BamHI site underlined) and 5′-GGATCCCCCCTGCTTGCACAGCTTCT-3′. The PCR products were cloned into a pGEM T vector (Promega), digested using BamHI and subcloned into a GFP fusion binary vector. The resulting CaMV 35S promoter-MPR25-GFP constructs and CaMV 35S promoter-F1F0ATPase targeting signal-RFP constructs (provided by Dr Shin-ichi Arimura, Graduate School of Life Sciences, Tokyo University, Japan) were co-bombarded into epidermal cells of Tulipa gesneriana leaves using a helium-driven accelerator (PDS/1000; Bio-Rad, http://www.bio-rad.com/). Bombardment parameters were as follows: 7585 kPa bombardment pressure, 1.0 μm gold particles, a distance of 6 cm from the macrocarrier to the samples, and a decompression vacuum of 94 kPa. After culture at room temperature for 1 day, the bombarded epidermal cells were viewed using a confocal scanning microscope system (Nikon, http://www.nikoninstruments.com/en_GB/).

To study the subcellular localization of MPR25 in stably transformed rice cells, a CaMV 35S promoter-MPR25-FlAG construct was created, replacing the GFP fragment of CaMV 35S promoter-MPR25-GFP by a Flag fragment from R4pGWB410 (Nakagawa et al., 2009). Transgenic rice plants expressing CaMV 35S promoter-MPR25-GFP were obtained using Agrobacterium-mediated transformation. Total mitochondrial proteins were purified from seed-derived calli by the method described previously (Tanaka et al., 2004; Kazama et al., 2008). Total chloroplast proteins were purified from 4-week-old seedlings using a chloroplast isolation kit (Sigma-Aldrich, http://www.sigmaaldrich.com/sigma-aldrich/home.html). Immunoblot analysis was performed as described previously (Kazama et al., 2008) using the following antibody concentrations: 1:1000 for anti-FLAG-alkarine phosphatase (Sigma-Aldrich), 1:5000 for anti- large subunit of ribulose-1,5-bisphosphate carboxylase/oxygenase (Agrisera, http://www.agrisera.com/), anti-isocitrate dehydrogenase (Agrisera) and alkaline phosphatase-conjugated anti-rabbit secondary antibody (Promega, http://www.promega.com/). Signals were detected using nitroblue tetrazolium chloride (Wako, http://www.wako-chem.co.jp/) and 5-bromo-4-chloro-3-indolyl phosphate p-toluidine salts (Wako).

Direct sequence analysis

Total RNA was isolated from 4-week-old etiolated seedlings and green leaves of 4-week-old seedlings using RNAiso (TaKaRa Bio). RNA (10 μg) was treated with DNase I (TaKaRa Bio), and the resulting RNA was reverse-transcribed using Rever Tra Ace (TOYOBO) and random primers (TaKaRa Bio). Mitochondrial and chloroplast editing sites were amplified using PCR primers as listed in Table S1. Total RNA was also isolated from root, mature pollen, calli and dry seeds, and used for direct sequence analysis of nad5. Amino acid sequences were obtained from an RNA editing database (REDIdb; http://biologia.unical.it/py_script/search.html) (Picardi et al., 2006). Amino acid alignment was performed using ClustalW implemented in Jalview (http://www.jalview.org/).

Northern blot analysis for nad5 transcript detection

Total RNA was isolated using RNAiso (TaKaRa Bio) from 4-week-old etiolated seedlings of WT and mpr25 mutants. RNA (10 μg) was subjected to Northern blot analysis as previously described (Kazama and Toriyama, 2003; Kazama et al., 2008). Probes labeled with digoxigenin (Roche, http://www.roche.com/) probes were obtained by PCR using primers 5′-ATACTTTTGGCTCTCGGGAG-3′ and 5′-TCCGTTGCAGAAAAGAGACC-3′.

Electrophoresis mobility shift assays

Expression and purification of recombinant MPR25 protein were performed as described by Kazama et al. (2008). The corresponding DNA sequence of the mature protein was amplified by PCR using primers 5′-AGATCTAAGAAGCTGTGCAAGCAGGG-3′ and 5′-AGATCTTCTATCCGTACTGTCGGCATC-3′ (BglII site underlined), and expressed using the pBAD/Thio-TOPO vector (Invitrogen, http://www.invitrogen.com/). The molecular size of the purified recombinant MPR25 protein (Trx-rMPR25) was confirmed by SDS–PAGE analysis as identical to the theoretical value of 84 kDa (Figure S7). The nad5 RNA probe corresponding to the 35 nucleotide proximal region (5′-AAAACUAAUACCUAUUCUGUUUAGUACUUCAGGUG-3′) of nad5eU1580SL (editing site is underlined) was synthesized by Operon Biotechnology, http://www.operonbiotech.com/, with digoxigenin-labeling at the 3′-end of the RNA. The nad7 RNA probe (5′-CUUUCCCAUGACGACUAGGAAAAGGCAAAUCAAAAAUUU-3′) used was identical to the sequence recognized by OTP87 (Hammani et al., 2011). EMSA was performed as previously described using 100 fmol recombinant MPR25 protein (Kazama et al., 2008).

Blue native PAGE and activity staining of NADH dehydrogenase

Mitochondrial protein was isolated from calli as previously described (Tanaka et al., 2004; Kazama et al., 2008). To isolate the mitochondrial membrane fraction, the pellet was suspended in a 1 m HEPES/KOH (pH 8.0), 0.3 m sorbitol suspension buffer. HEPES/KOH (50 mm, pH 7.5) was added to the suspended pellet, and the mixture was incubated on ice for 10 min. Then the mixture was centrifuged at 108 000 g for 1 h to precipitate the mitochondrial fraction. Blue native PAGE was performed using a NativePAGE sample prep kit (Invitrogen). The mitochondrial fraction was solubilized in Blue Native buffer (Invitrogen) supplemented with 1% w/v n-dodecyl-β-d-maltoside (DDM), and the mixture was incubated on ice for 15 min. The mixture was then centrifuged at 20 000 g for 30 min. The samples were supplemented with G-250 solution (Invitrogen). Dye-treated protein samples were directly loaded onto blue native gels. After blue native PAGE, the gel was stained with Coomassie Blue G-250 for 1 h, and de-stained in 10% v/v acetic acid to visualize the total protein loaded. For histochemical staining, the gel after blue native PAGE was stained in reaction buffer (0.1 m Tris/HCl pH 7.4, 0.2 mm NADH, 0.2% nitroblue tetrazolium) at room temperature.


This study was partially supported by a Grant-in-Aid (number 2338002) from the Ministry of Education, Science, Sports and Culture, Japan, and by the program for Promotion of Basic and Applied Researches for Innovations in Bio-oriented Industry. T.T. is the recipient of research fellowships from the Japan Society for the Promotion of Science for Young Scientists. We are grateful to Dr Akio Miyao and Dr Hirohiko Hirochika (National Institute of Agrobiological Science, Tsukuba, Japan) for kindly providing Tos17 line NC0057.