Silencing of the Nuclear RPS10 Gene Encoding MitochondrialRibosomal Protein Alters Translation inArabidopsis MitochondriaW

Malgorzata Kwasniak,a Pawel Majewski,a,1 Renata Skibior,a Aleksandra Adamowicz,a Malgorzata Czarna,a

Elwira Sliwinska,b and Hanna Janskaa,2

a Department of Biotechnology, University of Wroclaw, 51-148 Wroclaw, PolandbDepartment of Plant Genetics, Physiology, and Biotechnology, University of Technological and Life Sciences, 85-789 Bydgoszcz,Poland

ORCID ID: 0000-0002-6542-343X (HJ).

Hardly anything is known about translational control of plant mitochondrial gene expression. Here, we provide evidence fordifferential translation of mitochondrial transcripts in Arabidopsis thaliana. We found that silencing of the nuclear RPS10 geneencoding mitochondrial ribosomal protein S10 disturbs the ratio between the small and large subunits of mitoribosomes, withan excess of the latter. Moreover, a portion of the small subunits are incomplete, lacking at least the S10 protein. rps10 cellsalso have an increased mitochondrial DNA copy number per cell, causing an upregulation of all mitochondrial transcripts.Mitochondrial translation is also altered so that it largely overrides the hyperaccumulation of transcripts, and as a consequence,only ribosomal proteins are oversynthesized, whereas oxidative phosphorylation subunits are downregulated. Expression ofnuclear-encoded components of mitoribosomes and oxidative phosphorylation system (OXPHOS) complexes seems to be lessaffected. The ultimate coordination of expression of the nuclear and mitochondrial genomes occurs at the complex assemblylevel. These findings indicate that mitoribosomes can regulate gene expression by varying the efficiency of translation of mRNAsfor OXPHOS and ribosomal proteins.

INTRODUCTION

Bigenomic biogenesis is a characteristic feature of most mito-chondrial complexes, such as those of the oxidative phos-phorylation system (OXPHOS) and mitoribosomes. The stablestoichiometry of these complexes implies that the synthesis oftheir subunits must be tightly coordinated both within and be-tween the two genomes. The coordination between the nuclearand mitochondrial genomes could be achieved, in principle, atvarious levels (i.e., transcriptional, RNA maturation and degra-dation, translational, and posttranslational).

The regulation of expression of plant mitochondrially encodedgenes is not fully understood. Some results suggest a correla-tion between the number of mitochondrial gene copies and thesteady state level of their transcripts (Janska et al., 1998; Hedtkeet al., 1999; Kühn et al., 2009; Shedge et al., 2010), but there arealso data supporting a lack of such a correlation (Woloszynskaet al., 2006; Preuten et al., 2010). Differences in the gene pro-moter strength were observed between plant mitochondrialgenes (Brennicke et al., 1999; Fey and Maréchal-Drouard, 1999)

even for genes coding for components of the same complex(Giegé et al., 2000). However, the transcriptional activity appearsto have only a small effect on transcript abundance because it islargely counterbalanced by processes at different steps of RNAmaturation and degradation (Giegé et al., 2000; Leino et al.,2005; Holec et al., 2008). Therefore, it is thought that modulationof transcription is not essential for regulation of expression ofgenes encoded in mitochondria and that posttranscriptionalprocesses are more important. It should be emphasized thathardly anything is known about the regulation of plant mito-chondrial gene expression at the translational level (Bonen,2004). The importance of regulation at the posttranslational levelwas highlighted by the finding of selective proteolysis of pro-teins synthesized from unedited mRNA (Binder and Brennicke,2003) and by data demonstrating that the mitochondrial geneexpression remained more or less unaffected by sugar starva-tion at the transcriptional, posttranscriptional, and translationallevels, but was modulated at the posttranslational level (Giegéet al., 2005).Coordination of the expression of some OXPHOS genes be-

tween the nuclear and mitochondrial genomes at the transcriptlevel has been reported during, for example, flower developmentin sunflower (Helianthus annuus; Smart et al., 1994; Ribichichet al., 2001), wheat (Triticum aestivum) leaf development(Topping and Leaver, 1990), and early steps of rice (Oryza sativa)germination (Howell et al., 2009). On the other hand, an evidentlack of coordination between mitochondrial and nuclear tran-scripts of OXPHOS genes has been found in global analysesduring sugar starvation (Giegé et al., 2005) and germination inArabidopsis thaliana (Law et al., 2012). The sugar starvation

1Current address: Department of Biochemistry, Biophysics and Bio-technology, Jagiellonian University, Gronostajowa 7, 30-387 Krakow,Poland.2 Address correspondence to janska@ibmb.uni.wroc.pl.The author responsible for distribution of materials integral to the findingspresented in this article in accordance with the policy described in theInstructions for Authors (www.plantcell.org) is: Hanna Janska (janska@ibmb.uni.wroc.pl).W Online version contains Web-only data.www.plantcell.org/cgi/doi/10.1105/tpc.113.111294

The Plant Cell, Vol. 25: 1855–1867, May 2013, www.plantcell.org ã 2013 American Society of Plant Biologists. All rights reserved.

study has revealed that such coordination is achieved at theposttranslational level, possibly at the level of complex assem-bly (Giegé et al., 2005). Furthermore, that study suggested thatexcess unassembled subunits of the complexes were rapidlydegraded by ATP-dependent proteases, which had earlier beenidentified in plant mitochondria (Janska et al., 2010), and theirdeficit was shown to compromise the assembly/stability ofOXPHOS complexes (Kolodziejczak et al., 2007).

Despite the controversy over the transcriptional coordinationof mitochondrial and nuclear genes during the biogenesis ofmitochondrial complexes, the coordination of expression ofnuclear genes themselves is well documented. This regulationoperates through the interaction of nuclear transcription factorswith specific sequence motifs in promoter regions of nucleargenes encoding mitochondrial proteins. Numerous promotersof nuclear genes encoding OXPHOS components from Arabi-dopsis and rice contain a sequence known as site II motif(Welchen and Gonzalez, 2006), whose binding by TCP tran-scription factors ensures coordinated expression (Giraud et al.,2010; Hammani et al., 2011). Furthermore, a global study hasshown that expression of mitochondrially encoded genes in-volved in energy production and gene expression in mitochon-dria is also highly coregulated at the transcript level, even moretightly than are their nuclear counterparts (Leister et al., 2011).

In plants, all mitochondrial OXPHOS complexes, exceptcomplex II, comprise subunits encoded in both the mitochon-drial and the nuclear genome. In Arabidopsis, complex I containsthe largest proportion of mitochondrially encoded subunits (nineout of the total 30) (Heazlewood et al., 2003), while complex IIIcontains the least (one of 10) (Braun et al., 1994). Mitoribosomesalso have a dual genetic origin; however, their exact compositionis unknown. In Arabidopsis, the mitochondrial genome encodesfour ribosomal proteins of the small subunit (SSU), three of thelarge subunit (LSU), and all three rRNA species (Bonen, 2004). Thefull repertoire of nuclear-encoded mitoribosomal proteins remainsto be established.

One of the known components of the SSU of Arabidopsismitoribosomes encoded in the nucleus is the S10 protein.Transgenic lines with a decreased transcript level of ribosomalprotein S10 (RPS10) have been generated using the RNA in-terference method (Majewski et al., 2009). Homozygous Arabi-dopsis plants presented severe morphological abnormalitiesand either died during vegetative growth or were rescued byreversion. Among the hemizygous plants, several phenotypecategories were identified. These categories were correlatedwith the timing of the onset of silencing but not with its effi-ciency. The role of mitochondrial S10 protein in plants isunknown, but in bacteria its homolog, called as NusE, isa multifunctional protein (Kaczanowska and Rydén-Aulin, 2007).NusE is one of the proteins that complete the ribosomal as-sembly and is involved in transcriptional antitermination and intethering the ribosome to the RNA polymerase (Baumann et al.,2010).

In this work, we analyzed the effects of the silencing of theRPS10 gene encoding one of the SSU proteins on the expres-sion of mitochondrial and nuclear genes encoding constituentsof OXPHOS complexes and of mitoribosomes. Our findings re-veal that alterations in the mitoribosome divergently affect the

relative efficiency of translation of different types of mitochon-drial mRNA. As a consequence, the amount of mitochondriallyencoded ribosomal proteins is higher in rps10 plants than in thewild type, while the opposite is true for mitochondrially encodedOXPHOS subunits. These findings establish that differentialregulation of translation in plant mitochondria can be caused byalterations in mitoribosome status. We also confirmed usingplants rather than tissue culture the earlier suggestion of Giegéet al. (2005) that the coordination of expression of the mito-chondrial and nuclear genomes occurs at the complex assemblylevel.

RESULTS

Two Phenotypes Are Generated by Different Timing ofRPS10 Silencing

Two phenotypes of the rps10 Arabidopsis hemizygous trans-formants, P2 and P3, differing by the onset of RPS10 silencingwere reported earlier (Majewski et al., 2009). In the P2 pheno-type, the silencing occurs at the very early stage of vegetativegrowth and in consequence the decreased level of RPS10transcripts is observed in all rosette leaves. Since the silencingis associated with leaf morphological abnormalities, all leavesof P2 plants have an altered shape. By contrast, the de-velopmentally late silencing characteristic of the P3 phenotypebegins in newly emerging leaves of 7-week-old plants and re-sults in rosettes comprising leaves of different morphology; theolder ones are similar to the wild type, while the younger leavesresemble those of P2. An analysis of the unaltered and alteredleaves of P3 plants showed a correlation between the reducedRPS10 transcript level and leaf morphology at different de-velopmental stages of Arabidopsis (Figure 1). Unless notedotherwise, all analyses described below were performed usingthe youngest leaves harvested from 9- to 10-week-old P2, P3,and wild-type plants growing under short-day conditions. At that

Figure 1. Decreased Level of RPS10 Transcript Correlates with AlteredLeaf Morphology during Growth of P3 Phenotype of rps10.

The values obtained were averaged for three biological replicates, witherror bars representing SD. WT, the wild type.

1856 The Plant Cell

age, the young leaves of both P2 and P3 plants exhibited ab-normal morphology and decreased RPS10 transcript. It shouldbe emphasized that the efficiency of silencing is similar in thealtered leaves of P2 and P3 plants (Majewski et al., 2009).

Increased but Imbalanced Amounts of MitoribosomalSubunits Are Present in rps10 Plants

To check if the RPS10 silencing affects mitochondrial ribosomebiogenesis, the level of mitoribosomal subunits was quantifiedbased on rRNA abundance. As rRNAs are unstable when un-assembled, the rRNA level can serve as a proxy for ribosomalsubunit accumulation (Walter et al., 2010). We determined theabundance of cytosolic (cyt 18S and cyt 25S), mitochondrial (mt18S and mt 26S), and chloroplast (chl 16S and chl 23S) rRNAs(Figure 2). Since the S10 protein is a component of the SSU ofthe mitoribosome, a deficit of mitochondrial SSU and, conse-quently, also of 18S rRNA was expected in P2 and P3 plants.Surprisingly, we found an increased amount of mitochondrial18S rRNA in the mutant compared with the wild type (Figure 2A;the ratio mt 18S rRNA/cyt 18S rRNA was 1 for the wild type and;1.5 for both rps10 phenotypes). Moreover, the abundance ofthe mitochondrial rRNA of LSU, 26S rRNA, was even more in-creased (Figure 2A; the ratio mt 26S rRNA/cyt 25S rRNA was;3.5). We then calculated the ratio of LSU to SSU for the cy-tosolic, mitochondrial, and chloroplast ribosomes based on therespective rRNA levels (Figure 2B). A significant increase of thatratio was only found for mitoribosomes in both the P2 and P3phenotypes (Figure 2B; approximately twofold relative to the;1:1 stoichiometry expected and found for wild-type plants),

indicating an imbalance between mitoribosomal subunits. Thisconclusion is additionally supported by an approximately three-fold higher level of a polynucleotide phosphorylase-like protein,an enzyme known to be involved in 18S rRNA degradation(Perrin et al., 2004), in the P2 and P3 phenotypes (3.06 6 0.12and 2.78 6 0.05, respectively) compared with wild-type plants(1.00 6 0.06).To quantitate the abundance of the mitoribosomal subunits in

rps10 in an independent manner, we estimated the amounts ofthree mitoribosomal proteins, S4 and S10 of SSU and L16 rep-resenting LSU, using immunoblotting (Figure 3; see SupplementalFigure 1 online). S4 and L16 are encoded mitochondrially,whereas S10 is a nuclear-encoded protein. Both S4 and L16 weremore abundant in rps10 than in wild-type plants, but the increasewas substantially higher for L16 as calculated relative to totalmitochondrial protein. This result supports the above conclusionbased on rRNA quantity that the LSUs of mitoribosomes arein excess over SSUs in the rps10 plants. However, the amountof the S10 protein was comparable in the mitochondrial fraction ofrps10 and wild-type plants. This difference in the abundance ofthe S10 and S4 proteins suggests that in rps10 plants, a fractionof the mitochondrial SSUs were not fully assembled, lacking atleast the S10 protein.

OXPHOS Complexes and Their Subunits Are Less Abundantin rps10 Mitochondria

The effect of the increased but altered biogenesis of mito-ribosomes on the abundance and activity of OXPHOS complexeswas investigated by blue-native gel electrophoresis (BN-PAGE;Figure 4). The intensity of separated bands stained with Coo-massie blue (Figures 4A and 4C; protein level) or after specifichistochemical staining (Figures 4B and 4C; activity level) wasquantified. A significantly lower abundance of OXPHOS com-plexes was observed in rps10 compared with the wild type, bothas protein amount and activity level. The exception was complexIII that was reduced to a much lesser extent, while complex Iwas the most affected. This suggests an interesting correlationas complex III has only one and complex I the highest numberof mitochondrially encoded subunits among all OXPHOScomplexes.Subsequently, immunoblotting was performed using antibodies

against one mitochondrially encoded and one nuclear-encodedsubunit of complexes I, III, IV, and V (Figure 5; see SupplementalFigure 1 online). We consistently observed an approximatelysimilar reduction of the amount of mitochondrially and nuclear–encoded subunits for complex I (NAD9 and 49 kD; ;40% of wild-type level), III (COB and CYTC1; ;60% of wild-type level), and IV(COX2 and COXVC;;25% of wild-type level) but not for complexV. There, the level of a nuclear-encoded subunit was markedlyless affected (ATP2; ;50% of wild-type level) compared with amitochondrially encoded one (ATP6; ;15% of wild-type level).The evident imbalance between the above mitochondrially

and nuclear-encoded subunits of complex V prompted us tocheck the overall stoichiometry of the subunits of this complexusing two-dimensional BN-PAGE in the P2 phenotype of therps10 mutant. In contrast with the imbalance described above,the relative proportion of protein subunits of complex V was

Figure 2. Accumulation of rRNAs as a Proxy for Corresponding Ribo-somal Subunits in P2 and P3 Phenotypes of rps10 Compared with Wild-Type Plants.

(A) Ratio of mitochondrial to cytosolic rRNAs. WT, the wild type.(B) Ratio of rRNAs of large and SSUs in cytosolic, mitochondrial, andchloroplast ribosomes. The values obtained were averaged for four bi-ological replicates, with error bars representing SD. Statistically significantdifferences from the wild type are indicated by asterisks (Student’s t test;P < 0.05).

Translation in Arabidopsis Mitochondria 1857

nearly the same. The ratio of the four analyzed subunits ofcomplex V was as follows: ATP1 and ATP2/ATP g/FAd was1/0.19/0.10 in wild-type plants and 1/0.15/0.08 in P2 phenotypeof the rps10 mutant.

Mitochondrial DNA Content and Abundance ofMitochondrial mRNAs Are Upregulated in rps10 Plants

Since the biogenesis of OXPHOS complexes and mitoribosomesis bigenomic, we analyzed the consecutive steps of the expres-sion of the mitochondrial and nuclear genomes to understand theeffect of RPS10 silencing on the biogenesis of the mitochondrialcomplexes.

First, we investigated the abundance of mRNAs encodingrepresentative components of mitoribosomes and OXPHOScomplexes, both mitochondrially and nuclear encoded, usingquantitative RT-PCR (Figure 6). The abundance of mitochond-rially encoded transcripts was increased, while that of nuclear-encoded ones was unaltered or slightly decreased for all thecomplexes surveyed in rps10. The level of mitochondrial tran-scripts for most of the ribosomal proteins analyzed showeda somewhat stronger upregulation (twofold up to even eightfold)than did transcripts encoding the OXPHOS subunits (1.2-fold up

to fourfold). Although a similar trend was observed for bothphenotypes, the upregulation of the mitochondrially encodedgenes was more pronounced in P3 than in P2. Furthermore, theabundance of the majority of the nuclear-encoded transcriptsstudied tended to be unchanged in P3, but slightly decreased in

Figure 3. Quantification of Ribosomal Proteins in Mitochondrial Frac-tions from P2 and P3 Phenotypes of rps10 and Wild-Type Plants.

Protein gel blots of equal amounts (20, 10, and 5 mg) of total mito-chondrial proteins were probed with antibodies against S4, S10, andL16. Representative immunoblots are shown in Supplemental Figure 1online. Intensities of stained bands were analyzed with Image QuantSoftware and averaged for the three protein loadings. The values ob-tained were averaged for three independent experiments, with error barsrepresenting SD. Statistically significant differences from the wild type(WT) are indicated by asterisks (Student’s t test; P < 0.05).

Figure 4. Mitochondrial Respiratory Complexes Separated by BN-PAGEfrom P2 Phenotype of rps10 and Wild-Type Plants.

(A) Coomassie blue staining. WT, the wild type.(B) In-gel activity staining.(C) Quantification of the protein amount (top panel) and activity (bottompanel) of individual OXPHOS complexes in the P2 phenotype of rps10compared with wild-type plants. Intensities of stained bands were ana-lyzed with Image Quant Software. The values obtained were averaged forthree independent experiments, with error bars representing SD. Statis-tically significant differences from the wild type are indicated by asterisks(Student’s t test; P < 0.05).

1858 The Plant Cell

P2. We also noticed that the mitochondrial mRNAs encodingsubunits of complexes I and IV were more elevated (twofold upto fourfold) than those encoding subunits of complex III and V(2.5-fold or less) in the both phenotypes. This observation isinteresting in the light of the earlier finding that RPOTm, oneof the RNA polymerases present in Arabidopsis mitochondria,more efficiently transcribes genes encoding subunits of com-plexes I and IV (Kühn et al., 2009).

The fact that all of the mitochondrially encoded transcriptstested were elevated in rps10 prompted us to check if that in-crease could be a consequence of mitochondrial genome am-plification. To quantify the mitochondrial genome relative to thenuclear DNA content, we first established the ploidy of the nu-clear genome in leaf cells using flow cytometry. 2C, 4C, 8C, and16C nuclei were detected (see Supplemental Figure 2 online).Surprisingly, the mean C-values in P2 and P3 plants (5.02 60.42 and 4.34 6 0.47, respectively) were higher than in the wildtype (2.92 6 0.08; P = 0.05, Duncan’s test). This finding impliesthat the silencing of the RPS10 gene stimulated endore-duplication. Then, using quantitative PCR, we estimated DNAlevel for several mitochondrial genes encoding OXPHOS andribosomal proteins. These values were corrected for the nucleargenome endopolyploidization to give the absolute number ofmitochondrial gene copies per cell. Figure 7 presents copynumbers of individual mitochondrial genes in wild-type andrps10 cells expressed as log2 ratios. All of the mitochondrial

genes quantitated gave similar results: They were present in;4.5- to 6.5-fold more copies in the rps10 cells than in the wild-type ones. Given the above results (Figure 7), we reanalyzed thedata presented in Figure 6 to establish the relationship betweenthe copy number of mitochondrial genes and their RNA abun-dance. For several of the ribosomal proteins, the increase oftheir mRNA abundance reflected closely the increase in thegene copy number, implying that the accumulation of thosetranscripts was simply due the elevated mitochondrial DNA level.For the remaining ribosomal proteins, and all of the OXPHOSsubunits, the increase in abundance of their transcripts was notas high and more variable. Thus, the level of those mitochondrialtranscripts is most probably determined by a combined effect ofthe elevated cellular mtDNA level and the variable selectivity ofRNA polymerases.

Number of Mitochondria Is Unaffected in rps10 Mutants

The elevated number of mitochondrial gene copies per cellcould result from an increased DNA content per mitochondrionand/or an increased number of mitochondria per cell. To assessthe overall number of mitochondria per cell, we prepared pro-toplasts from P2, P3, and wild-type plants and counted themitochondria using MitoTracker GreenFM staining and confocalscanning microscopy. The mean number of mitochondria wasslightly higher in P2 and P3 compared with wild-type protoplasts

Figure 5. Mitochondrially and Nuclear-Encoded OXPHOS Proteins in Mitochondrial Fractions from P2 and P3 Phenotypes of rps10 Compared withWild-Type Plants.

Protein gel blots of equal amounts (20, 10, and 5 mg) of total mitochondrial proteins were probed with antibodies against the following: I, NAD9, 49 kD;III, COB, CYTC1; IV, COX2, COXVC; V, ATP6, ATP2. Representative immunoblots are shown in Supplemental Figure 1 online. Intensities of stainedbands were analyzed with Image Quant Software and averaged for the three protein loadings. The values obtained were averaged for three independentexperiments, with error bars representing SD. Statistically significant differences between mitochondrially and nuclear-encoded subunits within the samecomplex are indicated by asterisks (Student’s t test; P < 0.05). WT, the wild type.

Translation in Arabidopsis Mitochondria 1859

(see Supplemental Table 1 online), but that difference was notstatistically significant. Given the similar numbers of mitochon-dria in all three types of cells and the markedly different numbersof mitochondrial gene copies per cell, one must conclude thatmost if not all the rps10 mitochondria contain significantly moremitochondrial DNA copies than do wild-type mitochondria.

The Translation Status of OXPHOS and MitoribosomalmRNAs Is Differentiated in rps10 Plants

The perturbation of the assembly/stability of mitoribosomes inrps10 seemed likely to affect mitochondrial translation. To checkfor this, we determined the translational activity of individualmRNAs encoding OXPHOS and mitoribosomal proteins in rps10and wild-type plants. mRNA attached to ribosomes (polysomalfraction) was separated from unbound mRNA (nonpolysomalfraction) by Suc gradient centrifugation of total RNA extractsfrom P2, P3, and wild-type leaves, followed by quantitativeanalysis of specific transcripts by RT-PCR. Two parametersrelated to translatability of mRNA were analyzed: ribosomaldensity (the number of ribosomes present on a transcript visu-alized in the mRNA distribution across the polysome fraction)and ribosomal loading (the proportion of ribosome-boundmRNA expressed as the percentage of total mRNA). Repre-sentative data are shown in Figure 8 and Supplemental Figure 3online.

For the mitochondrially encoded cob, cox2, and atp6 tran-scripts, the mRNA distribution in the density gradient was shiftedtoward lighter fractions in both phenotypes of rps10 compared

with the wild type (Figure 8A). For the nad9 transcript, the shiftwas observed only for P3. These shifts indicate that the number ofribosomes present on the examined transcripts is lower in therps10 mutant, suggesting a lower efficiency of translation. Thesecond parameter describing translatability also implies a de-crease in mitochondrial translation for most of the OXPHOS

Figure 6. Steady State Transcript Levels of Mitochondrial and Nuclear Genes Encoding Subunits of OXPHOS and Mitoribosomes in P2 and P3Phenotypes of rps10 Compared with Wild-Type Plants.

The abundance of transcripts is expressed as a log2 ratio, in the following order: nad2, nad4, nad6, 76 kD, B14 SU, PSST SU, cob, CYTC1, 11 kD,MPPa, cox1, cox2, cox3, COXVC, COXVB, COXVIB, atp6, atp8, atp9, ATP2-1, ATP2-3, ATPd, rpl2, rpl5, rpl16, rps3, rps4, rps7, rps12, RPS14, andRPS24. The values obtained were averaged for four biological replicates, with error bars representing SD. WT, the wild type.

Figure 7. Mitochondrial Gene Copy Numbers in P2 and P3 Phenotypesof rps10 Compared with Wild-Type Plants.

The values were calculated per nucleus by multiplying the ratios of mi-tochondrial genes to nuclear DNA (ACT2 gene) by the mean C-value andare expressed as log2 ratios. The data were averaged for four biologicalreplicates, with error bars representing SD. Genes are presented in thefollowing order: I, nad9; III, cob; IV, cox2; V, atp6; mitoribosomes: rpl2,rpl5, rpl16, rps3, rps4, rps7, and rps12. WT, the wild type.

1860 The Plant Cell

transcripts, although the decrease in ribosomal loading wasconsiderably variable (Figure 8A; see Supplemental Figure 3Aonline). The only exception was the cob mRNA, showing a higherproportion in the polysomal fraction. The increased polysomeloading with the reduced ribosomal density found for the cobtranscript explains why the decrease of the COB protein amount

in rps10 was less drastic than that observed for the other mi-tochondrially encoded OXPHOS subunits (Figure 5). Conversely,enhanced translation was found for the majority of the mi-tochondrially encoded ribosomal transcripts as judged by botha shift to heavier polysomes in rps10 compared with the wildtype (rpl2, rpl5, rps4, and rps7; Figure 8A) and strongly enhanced

Figure 8. Distribution of mRNAs across Polysome Profiles for OXPHOS and Mitoribosomal Mitochondrial and Nuclear Genes.

(A) Transcripts of mitochondrially encoded genes (nad9, cob, cox2, atp6, rpl2, rpl5, rpl16, rps3, rps4, rps7, and rps12).(B) Transcripts of nuclear-encoded genes (76 kD, CYTC1, COXVC, ATP2-1, RPL29, RPS14, and RPS24). Specific mRNA was analyzed quantitativelyover the entire profile (fractions 1 to 10) by real-time PCR. Graphs represent percentage of total mRNA present in each fraction for wild-type (WT), P2,and P3 plants. The distribution of mRNA in the gradients was compared for P2 and P3 phenotypes of rps10 and wild-type plants, and shifts of theirmaxima to lighter or heavier gradient fractions relative to the wild-type maximum are indicated by arrows.

Translation in Arabidopsis Mitochondria 1861

loading onto polysomes (for all the mitochondrially encoded ri-bosomal mRNAs; Figure 8A; see Supplemental Figure 3A online).

A similar analysis of the nuclear-encoded transcripts revealedalmost unaffected cytoplasmic translation in the P2 phenotype(Figure 8B; see Supplemental Figure 3B online). In P3, a smalldecrease in ribosomal loading was accompanied by a higher ri-bosome density on translating mRNAs. Considering the contrastingchanges of ribosomal density and loading, we predict no sub-stantial changes in the translation efficiency of the nuclear-encodedOXPHOS and mitoribosomal transcripts in the P3 phenotype either.A clear exception was the COXVC mRNA showing a significantdecrease in polysome loading in both P2 and P3 phenotypes.

The Pattern of Proteins Synthesized in rps10 MitochondriaDiffers from That in the Wild Type

To provide an independent evidence for aberrant mitochondrialtranslation in rps10, we conducted in organello protein synthesisusing isolated mitochondria from the P2 and P3 rps10 mutantsand wild-type plants. Radioactively labeled mitochondrial pro-teins were separated by SDS-PAGE electrophoresis and visu-alized by autoradiography (Figure 9). As we expected from theaffected association of mRNAs with polysomes, the pattern oflabeled proteins synthesized by mitochondria of P2 and P3plants differed notably from the wild type. The profile of proteinssynthesized in mitochondria from wild-type plants resembledthat shown by Giegé et al. (2005) and Kühn et al. (2009). Bycontrast, several of those bands were hardly visible in P2 and P3mitochondria (Figure 9, open arrowhead), while some bands

barely detectable in wild-type mitochondria were very strong inP2 and P3 (Figure 9, closed arrowhead). These results confirmthat mitochondrial translation is misregulated in the rps10 mu-tant with enhanced synthesis of certain proteins and repressionof others. Based on the polysome analysis results, we postulatethat at least some of the proteins observed almost exclusivelyin P2 and P3 after the in organello translation are mitoribosomalproteins, while some of those hardly visible are OXPHOS sub-units. This prediction is supported by the data presented inSupplemental Figure 4 online. The apparent molecular massesof some proteins synthesized preferentially in mitochondria ofrps10 mutants or wild-type plants determined from theirelectrophoretic mobility matched well the theoretical sizesof several mitochondrially encoded ribosomal or OXPHOSproteins, respectively.

Mitochondrial ATP-Dependent Proteases Are Upregulatedin rps10 Plants

The data reported here indicate that only the efficiency of syn-thesis of mitochondrially encoded proteins is significantly alteredduring biogenesis of OXPHOS complexes in rps10 mutants, butnevertheless at the steady state protein level the ratio of bothmitochondrially and nuclear-encoded components is similar, atleast for the representative subunits of complexes I, III, and IV.This discrepancy suggests that the excess of synthesized pro-teins unassembled into the dual-origin complexes should beremoved. To test that prediction, we estimated the abundanceof mitochondrial ATP-dependent proteases known to digestunassembled/damaged proteins (Janska et al., 2010, 2013). Anincreased level of transcripts was detected only for the Lonproteases, LON1 and LON4 (Figure 10A), which degrade mainlysoluble proteins. Furthermore, despite unaltered transcript lev-els, the protein abundance of FtsH4 and FtsH10 proteases wasalso significantly increased in rps10 compared with wild-typemitochondria (Figure 10B). These proteases have been shown todegrade membrane proteins like OXPHOS subunits in yeast.

DISCUSSION

In this study, we have shown that a deficit in the ribosomalS10 protein leads to an imbalance between ribosome subunitsin Arabidopsis mitochondria. Furthermore, a portion of SSUs areincomplete, lacking at least the S10 protein. Next, we found thatthe expression of mitochondrially encoded components of mi-toribosomes and OXPHOS complexes is markedly modulated inresponse to the ensuing perturbations. As a consequence, theribosomal proteins are overexpressed, whereas the oxidativephosphorylation system subunits are downregulated. The ex-pression of nuclear-encoded components of mitoribosomes andOXPHOS complexes seems to be less affected. The final co-ordination of expression of the mitochondrial and nuclear genomesoccurs at the level of protein complex assembly, probably with aninvolvement of mitochondrial ATP-dependent proteases.To explain how the RPS10 silencing affects mitoribosome

biogenesis, we propose the following scenario. The silencingprovokes a compensatory response increasing the abundance

Figure 9. In Organello Protein Synthesis.

Autoradiogram of proteins synthesized for 30 and 60 min by mitochon-dria isolated from P2 and P3 phenotypes of rps10 and wild-type (WT)plants. After synthesis, 25 µg of mitochondria was fractionated by 12%(w/v) SDS-PAGE. The lane marked C. bact. (P3) is a specific controlperformed in mitochondria isolated from P3 in the presence of a sub-strate for bacterial translation. Open arrowheads, protein bands lessabundant in P2 and P3 than in wild-type plants; closed arrowheads,protein bands undetectable or less abundant in wild-type plants.

1862 The Plant Cell

of the components of both LSU and SSU of mitoribosomes.However, the upregulation is imbalanced since the S10-deficientSSU are unstable and therefore prone to degradation, resulting inan excess of LSU. It should be emphasized that owing to thefeedback response, the steady state level of the S10 protein isnearly the same in the mutant and wild-type mitochondria. Thus,the rps10 mutant does not suffer from an absolute insufficiencyof functional mitoribosomes but rather from the excess of LSUand/or the presence of a portion of partially assembled SSU. Thatsuch defective complexes are formed is indicated by an excess ofanother SSU component, S4, over S10. The goal of this studywas to evaluate how these changes in the mitoribosome statusaffect the multiple levels of the bigenomic biogenesis of mito-ribosomes themselves and oxidative phosphorylation complexes.

The first step of the feedback response in rps10 occurs at themitochondrial DNA level. A 4.5- to 6.5-fold increase in the mi-tochondrial gene copy number was detected in the P2 and P3

phenotypes. This increase was similar for all the genes tested,suggesting amplification of the whole mitochondrial genome.The observed upregulation was not accompanied by a pro-portional increase in the number/mass of mitochondria butwas reflected in enhanced steady state levels of mitochondrialtranscripts. The best correlation between the gene copy numberand the transcript level was found for several ribosomal proteins.Enhanced transcript levels as a consequence of elevated mi-tochondrial gene copy numbers have been reported earlier(Hedtke et al., 1999; Kühn et al., 2009; Shedge et al., 2010) andsuggested to reflect a feedback response to energy constraintscaused by defects in mitochondria (Kühn et al., 2009). In con-trast with the upregulated mitochondrial transcripts, the abun-dance of nuclear-encoded mRNAs of both mitoribosomes andOXPHOS complexes was unchanged or only slightly decreased.Thus, an evident lack of coordination between nuclear and mi-tochondrial gene expression at the transcript level was detectedin the rps10 mutant.The data regarding polysomes revealed substantial alterations

in mitochondrial translation in the rps10 plants. In contrast withthe transcript level, where the same trend was noted for bothmitoribosome and OXPHOS proteins, a divergent translationalresponse was observed. Generally, within mitochondria, themajority of transcripts of OXPHOS subunits are less activelytranslated, whereas transcripts of most mitoribosome proteinsare preferentially synthesized. This contrasting response waspreserved at the protein steady state level, where the OXPHOSsubunits were present at reduced levels, while ribosomal pro-teins overaccumulated. The translation efficiency of nuclear-encoded mRNAs for mitoribosome and OXPHOS componentswas much less affected. However, there was a significant re-duction in the steady state protein level of the nuclear-encodedOXPHOS subunits of complexes I, III, and IV. Moreover, thisdecrease was comparable to that found for the mitochondriallyencoded components of the same complexes. Thus, it seemsthat the nuclear-encoded OXPHOS proteins, after almost un-changed synthesis, are degraded to bring their abundancedown to the level of the mitochondrially encoded components.No similar analysis could be performed for mitoribosomes dueto a lack of antibodies against their nuclear-encoded constituents,except for S10.The conclusion that the coordination of expression of mito-

chondrial and nuclear genes coding for subunits of OXPHOScomplexes occurs posttranslationally was reached earlier byGiegé et al. (2005) in their study of the biogenesis of ArabidopsisOXPHOS complexes under sugar starvation conditions. Theypostulated that the coordination takes place at the level ofprotein complex assembly. Our results further confirmed thatsuggestion. The nuclear-encoded subunit of complex V wasreduced in rps10 to a markedly lesser extent than its mi-tochondrially encoded one, but the stoichiometry of the complexV subunits was virtually identical to that in wild-type mitochon-dria. Apart from this similarity, there are also differences in thefeedback responses initiated by sugar starvation and by RPS10silencing. The response to the RPS10 silencing is mainly as-sociated with changes in mitochondrial genome expression,whereas under sugar starvation the biogenesis of OXPHOScomplexes was regulated mostly by changes in nuclear genome

Figure 10. Abundance of Mitochondrial ATP-Dependent Proteases inP2 and P3 Phenotypes of rps10 and Wild-Type Plants.

(A) Steady state transcript level. The values obtained were averaged forfour biological replicates, with error bars representing SD. WT, the wildtype.(B) Steady state protein level. Protein gel blots of equal amounts (20, 10,and 5 mg) of total mitochondrial proteins were probed with antibodiesagainst FtsH4 and FtsH10. Quantification of intensities of stained bandswas done with Image Quant Software and averaged for the three proteinloadings. Values obtained were averaged for three independent experi-ments, with error bars representing SD. Statistically significant differencesfrom the wild type are indicated by asterisks (Student’s t test; P < 0.05).

Translation in Arabidopsis Mitochondria 1863

expression. As a consequence, the limiting factor in the bio-genesis of OXPHOS complexes in response to sugar deprivationis the availability of nuclear-encoded subunits, while in the caseof RPS10 silencing it is the availability of mitochondrially en-coded ones. Our results argue that the excess unassembledmitochondrial proteins are degraded by ATP-dependent pro-teases. We found transcriptional activation of the Lon1 and Lon4proteases in both phenotypes of rps10 and upregulation ofFtsH4 and FtsH10 proteases at the posttranscriptional level. Thepossibility of translational regulation of FtsH10 expression hadbeen reported earlier (Piechota et al., 2010).

Although the molecular details remain to be determined, ourdata indicate that the altered translation found in rps10 mito-chondria is caused by changes of at least two factors: ribosomalloading and density. The reduced ribosomal densities found forall the mitochondrial OXPHOS transcripts tested explains atleast partially the reduced steady state levels of the corre-sponding OXPHOS subunits. It seems that the diverse decreasein the steady state levels of individual OXPHOS proteins reflectsthe differences in the ribosomal loading of their mRNAs. Con-versely, the highly increased steady state levels of mitochondriallyencoded ribosomal proteins result from enhanced ribosomalloading and for majority of proteins also from increased ribo-somal density. Taken together, the perturbation in mito-ribosomes resulting from RPS10 silencing leads to extensiveand diverse changes in mitochondrial translation reflected in thesteady state protein level. The mitoribosomal proteins are over-accumulated, while the OXPHOS subunits are present at reducedlevels in rps10. This demonstrates that differential changes of theefficiency of translation in response to compromised ribosomalassembly/stability occur in plant mitochondria. Autoregulation ofribosome biosynthesis by translational control has been reportedfor a fission yeast mutant depleted of Arg-methylated ribosomalprotein S2 (Bachand et al., 2006). This mutant, like the Arabi-dopsis rps10, shows an imbalanced level of small and large ri-bosomal subunits. However, in the yeast mutant, an enhancedtranslation of several mRNAs encoding proteins of SSU wasobserved with little variation in the overall mRNA abundance.Translational regulation of ribosome biogenesis has also beendocumented in bacteria (Nomura, 1999), mammalian cells(Meyuhas and Hornstein, 2000), and chloroplasts (Fleischmannet al., 2011).

Coupling of the rate of synthesis of organellar proteins to therate of their assembly into complexes is a common feature ofseveral related mechanisms known as control by epistasy ofsynthesis (CES) (Choquet and Vallon, 2000). In this context, itis feasible that a CES-like regulation triggered by partially as-sembled mitoribosome may operate to activate translation ofribosomal proteins in rps10 mitochondria. However, the regu-lation of translation by the assembly status of the mitoribosomeobserved in the rps10 mutant is not limited to the componentsof the mitoribosome itself, as would be postulated by the CESmodel, but also concerns the OXPHOS subunits, being thereforemore general. The effect of the accumulation of free LSU and thepresence of a portion of defective SSU on the efficiency ofprotein synthesis in plant mitochondria could be based on theformation of atypical interactions between those subunits andother components of the translation apparatus. The ribosome

filter hypothesis posits that ribosomes are not simply translationmachines but also function as regulatory elements that differ-entially affect or filter the translation of particular mRNAs (Mauroand Edelman, 2007). Our findings support the hypothesis in thatthe misregulation of the mitoribosome subunits in rps10 differ-entially affects the translational status of particular mitochondrialtranscripts. We further postulate that similar translational regu-lation could occur in wild-type plants. Variation in the inter-actions between the ribosome subunits could be triggered byposttranslational modifications of ribosomal proteins. In thisrespect, it is interesting to note that methylation of the S10protein in bacteria affects the assembly of ribosomes and pro-tein synthesis (Ren et al., 2010).Surprisingly, we found that whereas most nuclei in wild-type

plants were diploid (mean C-value ;3), the mutants nuclei ex-hibited a higher ploidy (mean C-value ;4 to 5). However, thischange is not dramatic in the light of data showing that the meanC-value of nuclear DNA varies from 4 up to 13 during maturationand aging of Arabidopsis leaves (Zoschke et al., 2007). Recently,it was shown that the mammalian homolog of the S10 protein,like many other ribosomal proteins, has a secondary function asa cell proliferation regulator (Ren et al., 2010; Abbas et al., 2012).Thus, it is highly likely that the plant S10 also has an extraribosomalfunction connected with cell proliferation and a deficiency of thisprotein leads to endoreduplication. Further studies will be neededto test this hypothesis.

METHODS

Plant Material and Growth Conditions

Wild-type and mutant Arabidopsis thaliana plants were of the Columbia-0ecotype. The transgenic lines of rps10 were generated using the RNAinterference method (Majewski et al., 2009). Seeds of wild-type plantswere germinated on half-strength Murashige and Skoog, supplementedwith kanamycin (50 mg L21) for the rps10 mutant. Seedlings were grownfor 3 d at 4°C in the dark followed by 11 d in an 8-h-light/16-h-dark (shortday) photoperiod at 22°C with light intensity of 100 mmol m22 s21. Afterthat time, the wild-type and the kanamycin-resistant rps10 plants weretransferred to the soil and grown in a growth chamber (short day, 22°C,100 mmol m22 s21).

Nucleic Acid Isolation and cDNA Synthesis

Nucleic acids were extracted from fresh plant leaves and stored at –20°C.Genomic DNA was isolated using a GeneMATRIX Plant and Fungi DNApurification kit (EURx). Total RNA was isolated using a GeneMATRIXUniversal RNA purification kit (EURx). The reverse transcription reactionwas performed using up to 2 mg of total RNA and a reverse transcriptionkit (Applied Biosystems). The resulting cDNA was used as a template forquantitative real-time PCR.

Protoplast Isolation, Confocal Imaging, and Determination of theNumber of Mitochondria per Cell

Protoplasts were isolated from leaves using the method described by Yooet al. (2007). Isolated protoplasts were pelleted at 1000g for 1 min at roomtemperature and resuspended in 50 mL MMg solution (0.4 M mannitol,15 mM MgCl2, and 4 mM MES, pH 5.7) containing 200 nM MitoTrackerGreenFM (Molecular Probes), prewarmed at 37°C. Images of a half-

1864 The Plant Cell

protoplast were acquired as a z-series with 0.39-µm intervals usinga Zeiss LSM510 confocal laser scanning microscope equipped witha 360 water immersion objective. The MitoTracker GreenFM stain wasvisualized with a 488-nm argon laser and a BP505-530 filter. The numberof mitochondria was estimated using an approach similar to that ofPreuten et al. (2010). Twenty randomly chosen protoplasts from eachof three separate isolations were analyzed for P2 and P3 phenotypes ofrps10 and for wild-type plants. The number of mitochondria per protoplastwas determined using Imaris Software 7.6.1 (Bitplane).

Flow Cytometric Analysis of Nuclear DNA Content

For flow cytometry, sampleswere prepared as described previously (Sliwinskaet al., 2012). Single leaves were chopped in 1 mL of nuclei isolation buffersupplemented with propidium iodide (50 mg mL21) and RNase A (50 mgmL21) and analyzed using a CyFlow SL Green flow cytometer (Partec)equipped with a high-grade solid-state laser with green light emission at532 nm, long-pass filter RG590 E,DM560 A, and side and forward scatters.For each sample, the DNA content of 3000 to 7000 nuclei was measured.Analyses were performed on at least six replicates using a logarithmicamplification. Histograms were evaluated using the FloMax program(Partec), and the percentage of nuclei with particular DNA contents andthe mean C-value (Lemontey et al., 2000) were calculated before applyinga one-way analysis of variance and the Duncan’s test.

Isolation of Polysomal RNA

Polysomes were fractionated from homogenized leaf tissue as describedpreviously by Kahlau and Bock (2008) with slight modifications. Leaves(200mg) were ground in 1mL of freshly prepared extraction buffer (200mMTris-HCl, pH 9.0, 200 mM KCl, 35 mM MgCl2, 25 mM EGTA,200 mMSuc, 100 mM b-mercapthoethanol, 1% [v/v] Triton X-100, 2% [v/v]polyoxyethylene-10-tridecyl ether, 6% [w/v] digitonin, 1 mg/mL heparin,100 mg/mL chloramphenicol, and 25 mg/mL cycloheximide) and centri-fuged at 4°C and 13,200g for 5 min to remove cell debris. The clearedextract was supplemented with 1/20 volume of 10% (w/v) sodium de-oxycholate and fractionated in a Suc density gradient (56:40:30:15% [w/v]Suc in 40 mM Tris-HCl, pH 8.5, 20 mM KCl, 10 mM MgCl2, 100 mg/mLchloramphenicol, and 500 mg/mL heparin) by ultracentrifugation (4°C,200,000g, 80 min). Ten aliquots of ;410 mL each obtained by fraction-ation were subjected to RNA extraction using phenol/chloroform/isoamylalcohol (25:24:1). Control gradients supplemented with puromycin wereused to identify fractions containing polysomes (fractions 6 to 10) and freemRNA and ribosomes (fractions 1 to 5). RNA isolated from all of thesefractions was subjected to 2.5 M LiCl treatment to remove the heparin.Reverse transcription was performed as described above and cDNA usedfor quantitative real-time PCR.

Real-Time PCR Analysis

The real-time PCR analyses were performed using DNA or cDNA ona LightCycler 2.0 instrument (Roche Applied Science). The Real-Time2x PCR Master Mix SYBR version B (A&A Biotechnology) was used.Reactions were performed in a total volume of 15 mL with a final con-centration of 0.5 mM primers. Relative quantification analysis using thesecond derivative maximum method of the LightCycler 4.0 software wasused. The wild-type plants served as the calibrator, and the ACT2 gene(At3g18780) was used as a reference. The values of amplification effi-ciency of the analyzed amplicons were calculated on a standard curve.The standard curves were generated using six serial twofold dilutions ofthe cDNA or DNA samples obtained from wild-type plants. The amplifi-cation protocol consisted of the following: denaturation, 95°C for 1 min;amplification, 45 cycles at 95°C for 10 s, 50 to 65°C (the annealingtemperature was specific for used primers) for 10 s, 72°C for 20 s with

single data acquisition; cooling, 40°C for 30 s. The specificity of theamplification products was verified by analysis of the melting curve. Theprimers used are listed in Supplemental Table 2 online.

Isolation of Mitochondria and Gel Electrophoresis of Proteins

Isolation of mitochondria for SDS-PAGE and BN-PAGE experiments wasperformed as described by Urantowka et al. (2005), whereas functionalmitochondria for in organello protein synthesis were isolated as describedby Lister et al. (2007). SDS-PAGE electrophoresis as well as protein andcatalytic staining of BN-PAGE gels were performed as described byKolodziejczak et al. (2007).

Protein Gel Blot Analysis

Equal amounts of mitochondrial proteins (20, 10, and 5 mg) were frac-tionated by SDS-PAGE and transferred to a polyvinylidene difluoridemembrane (Bio-Rad). Blots were probed with antibodies against the fol-lowing OXPHOS subunits: NAD9 (obtained from P. Giegé, France), 49 kD(obtained from L. Sweetlove, UK), COB (obtained from L. Sweetlove),CYTC1 (obtained from P. Giegé), COX2 (purchased from Agrisera), COXVC(obtained from P. Giegé), ATP6 (obtained from Christiane D. Chase, USA),andATP2 (obtained fromC. Knorpp, Sweden). Furthermore, blots were alsoprobed with antibodies purchased from Agrisera, against Arabidopsis mi-toribosomal proteins, anti-S10, anti-S4, and anti-L16, as well as againstmitochondrial proteases, anti-FtsH4 andanti-FtsH10. Anti-rabbit antibodiesconjugated with horseradish peroxidase were used as secondary an-tibodies and visualized with enhanced chemiluminescent ECL reagent.Chemiluminescence was recorded using a chemiluminescence imager(BioDoc-It Imaging System; UVP). Bands were quantified using Quanti-tyOne software (Bio-Rad).

In Organello Protein Synthesis

Mitochondrial proteins (150 mg) were resuspended in a solution con-taining 5 mM KH2PO4, pH 7.0, 400 mM mannitol, 60 mM KCl, 50 mMHEPES, 10 mM MgCl2, 10 mM malic acid, 1 mM pyruvate, 2 mM GTP,2 mM DTT, 4 mM ADP, 0.1% (w/v) BSA, 25 mM unlabeled 19–amino acidsolution, and 30mCi [35S]Met (>1000Ci/mmol). Reactions were performedin 100 mL for 30 and 60 min at 25°C on an orbital shaker and stopped byaddition of puromycin (50 mg/mL) and 350 mL mitochondria wash buffercontaining 10 mM unlabeled Met. In control experiments, 25 mMNa-acetate was used instead of malic acid and pyruvate. Radio-labeled proteins were separated by SDS-PAGE and visualized byautoradiography.

Accession Numbers

Sequence data from this article can be found in the Arabidopsis GenomeInitiative or GenBank/EMBL databases under the following accessionnumbers: AtMg00090 (rps3), AtMg00290 (rps4), AtMg01270 (rps7),At3g22300 (RPS10), AtMg00980 (rps12), At2g34520 (RPS14), At5g28060(RPS24), AtMg00560 (rpl2), AtMg00210 (rpl5), AtMg00080 (rpl16),At1g07830 (RPL29), AtMg00285 (nad2), AtMg00580 (nad4), AtMg00270(nad6), AtMg00070 (nad9), At5g37510 (76 kD), At3g12260 (B14 SU),At5g11770 (PSST SU), AtMg00220 (cob), At5g40810 (CYTc1), At1g15120(11 kD SU), At1g51980 (MPP a), AtMg01360 (cox1), AtMg00160 (cox2),AtMg00730 (cox3), At1g80230 (COXVb), At2g47380 (COXVc), At4g28060(COXVIb), AtMg01170 (atp6), AtMg00480 (atp8), AtMg01080 (atp9),At5g08670 (ATP2-1), At5g08690 (ATP2-3), At2g33040 (ATPg), At5g26860(LON1), At3g05790 (LON4), At5g23140 (CLPP2), At5g53350 (CLPX ),At2g29080 (FTSH3), At2g26140 (FTSH4), At1g07510 (FTSH10), andAt3g18780 (ACT2).

Translation in Arabidopsis Mitochondria 1865

Supplemental Data

The following materials are available in the online version of this article.

Supplemental Figure 1. Representative Immunoblots of Total Mito-chondrial Protein Isolated from Wild-Type, P2, and P3 Phenotypes ofrps10 Plants.

Supplemental Figure 2. Selected DNA Histograms of NuclearPreparations from Wild-Type Plants and P2 and P3 Phenotypes ofrps10.

Supplemental Figure 3. Ribosome Loading for Mitochondrially andNuclear-Encoded Gene Transcripts in P2 and P3 Phenotypes of rps10and Wild-Type Plants.

Supplemental Figure 4. Identification of Mitochondrially EncodedProteins Whose Expression Is Affected by the Mitoribosome Defect.

Supplemental Table 1. Number of Mitochondria per Cell.

Supplemental Table 2. Primers Used for qRT-PCR.

ACKNOWLEDGMENTS

We thank Christine Chase, Carina Knorpp, Philippe Giegé, and LeeSweetlove for supplying the antibodies for OXPHOS complexes. Thiswork was supported by Grant N N301 784940 from the Ministry ofScience and Higher Education, Poland.

AUTHOR CONTRIBUTIONS

H.J. designed the study. M.K., P.M., R.S., A.A., M.C., and E.S. performedexperiments. H.J., M.K., R.S., A.A., M.C., and E.S. analyzed data. H.J.wrote the article with significant input from M.K.

Received March 7, 2013; revised May 3, 2013; accepted May 10, 2013;published May 30, 2013.

REFERENCES

Abbas, W., Dichamp, I., and Herbein, G. (2012). The HIV-1 Nefprotein interacts with two components of the 40S small ribosomalsubunit, the RPS10 protein and the 18S rRNA. Virol. J. 9: 103.

Bachand, F., Lackner, D.H., Bähler, J., and Silver, P.A. (2006).Autoregulation of ribosome biosynthesis by a translational responsein fission yeast. Mol. Cell. Biol. 26: 1731–1742.

Baumann, M., Pontiller, J., and Ernst, W. (2010). Structure and basaltranscription complex of RNA polymerase II core promoters in themammalian genome: An overview. Mol. Biotechnol. 45: 241–247.

Binder, S., and Brennicke, A. (2003). Gene expression in plantmitochondria: Transcriptional and post-transcriptional control. Philos.Trans. R. Soc. Lond. B Biol. Sci. 358: 181–188, discussion 188–189.

Bonen, L. (2004). Translational machinery in plant organelles. In MolecularBiology and Biotechnology of Plant Organelles, H. Daniell and C. Chase,eds (Dordrecht, The Netherlands: Springer), pp. 323–345.

Braun, H.P., Kruft, V., and Schmitz, U.K. (1994). Molecularidentification of the ten subunits of cytochrome-c reductase frompotato mitochondria. Planta 193: 99–106.

Brennicke, A., Zabaleta, E., Dombrowski, S., Hoffmann, M., andBinder, S. (1999). Transcription signals of mitochondrial andnuclear genes for mitochondrial proteins in dicot plants. J. Hered.90: 345–350.

Choquet, Y., and Vallon, O. (2000). Synthesis, assembly and degradationof thylakoid membrane proteins. Biochimie 82: 615–634.

Fey, J., and Maréchal-Drouard, L. (1999). Expression of the twochloroplast-like tRNA(Asn) genes in potato mitochondria: Mapping oftranscription initiation sites present in the trnN1-trnY-nad2 cluster andupstream of trnN2. Curr. Genet. 36: 49–54.

Fleischmann, T.T., Scharff, L.B., Alkatib, S., Hasdorf, S., Schöttler,M.A., and Bock, R. (2011). Nonessential plastid-encodedribosomal proteins in tobacco: A developmental role for plastidtranslation and implications for reductive genome evolution. PlantCell 23: 3137–3155.

Giegé, P., Hoffmann, M., Binder, S., and Brennicke, A. (2000). RNAdegradation buffers asymmetries of transcription in Arabidopsismitochondria. EMBO Rep. 1: 164–170.

Giegé, P., Sweetlove, L.J., Cognat, V., and Leaver, C.J. (2005).Coordination of nuclear and mitochondrial genome expressionduring mitochondrial biogenesis in Arabidopsis. Plant Cell 17:1497–1512.

Giraud, E., Ng, S., Carrie, C., Duncan, O., Low, J., Lee, C.P., VanAken, O., Millar, A.H., Murcha, M., and Whelan, J. (2010). TCPtranscription factors link the regulation of genes encoding mitochondrialproteins with the circadian clock in Arabidopsis thaliana. Plant Cell 22:3921–3934.

Hammani, K., Gobert, A., Hleibieh, K., Choulier, L., Small, I., andGiegé, P. (2011). An Arabidopsis dual-localized pentatricopeptiderepeat protein interacts with nuclear proteins involved in geneexpression regulation. Plant Cell 23: 730–740.

Heazlewood, J.L., Howell, K.A., and Millar, A.H. (2003). Mitochondrialcomplex I from Arabidopsis and rice: Orthologs of mammalian and fungalcomponents coupled with plant-specific subunits. Biochim. Biophys.Acta 1604: 159–169.

Hedtke, B., Wagner, I., Börner, T., and Hess, W.R. (1999). Inter-organellar crosstalk in higher plants: Impaired chloroplast developmentaffects mitochondrial gene and transcript levels. Plant J. 19: 635–643.

Holec, S., Lange, H., Dietrich, A., and Gagliardi, D. (2008).Polyadenylation-mediated RNA degradation in plant mitochondria.Methods Enzymol. 447: 439–461.

Howell, K.A., Narsai, R., Carroll, A., Ivanova, A., Lohse, M., Usadel,B., Millar, A.H., and Whelan, J. (2009). Mapping metabolic andtranscript temporal switches during germination in rice highlightsspecific transcription factors and the role of RNA instability in thegermination process. Plant Physiol. 149: 961–980.

Janska, H., Kwasniak, M., and Szczepanowska, J. (2013). Proteinquality control in organelles - AAA/FtsH story. Biochim. Biophys.Acta 1833: 381–387.

Janska, H., Piechota, J., and Kwasniak, M. (2010). ATP-dependentproteases in biogenesis and maintenance of plant mitochondria.Biochim. Biophys. Acta 1797: 1071–1075.

Janska, H., Sarria, R., Woloszynska, M., Arrieta-Montiel, M., andMackenzie, S.A. (1998). Stoichiometric shifts in the common beanmitochondrial genome leading to male sterility and spontaneousreversion to fertility. Plant Cell 10: 1163–1180.

Kaczanowska, M., and Rydén-Aulin, M. (2007). Ribosome biogenesisand the translation process in Escherichia coli. Microbiol. Mol. Biol. Rev.71: 477–494.

Kahlau, S., and Bock, R. (2008). Plastid transcriptomics andtranslatomics of tomato fruit development and chloroplast-to-chromoplast differentiation: Chromoplast gene expression largelyserves the production of a single protein. Plant Cell 20: 856–874.

Kolodziejczak, M., Gibala, M., Urantowka, A., and Janska, H.(2007). The significance of Arabidopsis AAA protease for activityand assembly/stability of mitochondrial OXPHOS complexes. Physiol.Plant. 129: 135–142.

1866 The Plant Cell

Kühn, K., Richter, U., Meyer, E.H., Delannoy, E., de Longevialle,A.F., O’Toole, N., Börner, T., Millar, A.H., Small, I.D., and Whelan,J. (2009). Phage-type RNA polymerase RPOTmp performs gene-specific transcription in mitochondria of Arabidopsis thaliana. PlantCell 21: 2762–2779.

Law, S.R., Narsai, R., Taylor, N.L., Delannoy, E., Carrie, C., Giraud,E., Millar, A.H., Small, I., and Whelan, J. (2012). Nucleotide andRNA metabolism prime translational initiation in the earliest eventsof mitochondrial biogenesis during Arabidopsis germination. PlantPhysiol. 158: 1610–1627.

Leino, M., Landgren, M., and Glimelius, K. (2005). Alloplasmiceffects on mitochondrial transcriptional activity and RNA turnoverresult in accumulated transcripts of Arabidopsis orfs in cytoplasmicmale-sterile Brassica napus. Plant J. 42: 469–480.

Leister, D., Wang, X., Haberer, G., Mayer, K.F., and Kleine, T.(2011). Intracompartmental and intercompartmental transcriptionalnetworks coordinate the expression of genes for organellar functions.Plant Physiol. 157: 386–404.

Lemontey, C., Mousset-Déclas, C., Munier-Jolain, N., and Boutin,J.P. (2000). Maternal genotype influences pea seed size by controlling bothmitotic activity during early embryogenesis and final endoreduplicationlevel/cotyledon cell size in mature seed. J. Exp. Bot. 51: 167–175.

Lister, R., Carrie, C., Duncan, O., Ho, L.H., Howell, K.A., Murcha,M.W., and Whelan, J. (2007). Functional definition of outermembrane proteins involved in preprotein import into mitochondria.Plant Cell 19: 3739–3759.

Majewski, P., Wołoszy�nska, M., and Ja�nska, H. (2009). Developmentallyearly and late onset of Rps10 silencing in Arabidopsis thaliana: Geneticand environmental regulation. J. Exp. Bot. 60: 1163–1178.

Mauro, V.P., and Edelman, G.M. (2007). The ribosome filter redux.Cell Cycle 6: 2246–2251.

Meyuhas, O., and Hornstein, E. (2000). Translational control of TOPmRNAs. In Translational Control of Gene Expression, N. Sonenberg,J.W.B. Hershey, and M.B. Mathews, eds (Cold Spring Harbor, NY:Cold Spring Harbor Laboratory Press), pp. 671–693.

Nomura, M. (1999). Regulation of ribosome biosynthesis in Escherichiacoli and Saccharomyces cerevisiae: Diversity and common principles.J. Bacteriol. 181: 6857–6864.

Perrin, R., Lange, H., Grienenberger, J.M., and Gagliardi, D. (2004).AtmtPNPase is required for multiple aspects of the 18S rRNAmetabolism in Arabidopsis thaliana mitochondria. Nucleic AcidsRes. 32: 5174–5182.

Piechota, J., Kolodziejczak, M., Juszczak, I., Sakamoto, W., andJanska, H. (2010). Identification and characterization of highmolecular weight complexes formed by matrix AAA proteases andprohibitins in mitochondria of Arabidopsis thaliana. J. Biol. Chem.285: 12512–12521.

Preuten, T., Cincu, E., Fuchs, J., Zoschke, R., Liere, K., andBörner, T. (2010). Fewer genes than organelles: Extremely low andvariable gene copy numbers in mitochondria of somatic plant cells.Plant J. 64: 948–959.

Ren, J., Wang, Y., Liang, Y., Zhang, Y., Bao, S., and Xu, Z. (2010).Methylation of ribosomal protein S10 by protein-arginine methyltransferase5 regulates ribosome biogenesis. J. Biol. Chem. 285: 12695–12705.

Ribichich, K.F., Tioni, M.F., Chan, R.L., and Gonzalez, D.H. (2001).Cell-type-specific expression of plant cytochrome c mRNA in developingflowers and roots. Plant Physiol. 125: 1603–1610.

Shedge, V., Davila, J., Arrieta-Montiel, M.P., Mohammed, S., andMackenzie, S.A. (2010). Extensive rearrangement of the Arabidopsismitochondrial genome elicits cellular conditions for thermotolerance.Plant Physiol. 152: 1960–1970.

Sliwinska, E., Mathur, J., and Bewley, J.D. (2012). Synchronouslydeveloping collet hairs in Arabidopsis thaliana provide an easilyaccessible system for studying nuclear movement and endoreduplication.J. Exp. Bot. 63: 4165–4178.

Smart, C.J., Monéger, F., and Leaver, C.J. (1994). Cell-specificregulation of gene expression in mitochondria during anther developmentin sunflower. Plant Cell 6: 811–825.

Topping, J.F., and Leaver, C.J. (1990). Mitochondrial gene expressionduring wheat leaf development. Planta 182: 399–407.

Urantowka, A., Knorpp, C., Olczak, T., Kolodziejczak, M., andJanska, H. (2005). Plant mitochondria contain at least two i-AAA-like complexes. Plant Mol. Biol. 59: 239–252.

Walter, M., Piepenburg, K., Schöttler, M.A., Petersen, K., Kahlau, S.,Tiller, N., Drechsel, O., Weingartner, M., Kudla, J., and Bock, R.(2010). Knockout of the plastid RNase E leads to defective RNA processingand chloroplast ribosome deficiency. Plant J. 64: 851–863.

Welchen, E., and Gonzalez, D.H. (2006). Overrepresentation of elementsrecognized by TCP-domain transcription factors in the upstream regionsof nuclear genes encoding components of the mitochondrial oxidativephosphorylation machinery. Plant Physiol. 141: 540–545.

Woloszynska, M., Kmiec, B., Mackiewicz, P., and Janska, H.(2006). Copy number of bean mitochondrial genes estimated byreal-time PCR does not correlate with the number of gene loci andtranscript levels. Plant Mol. Biol. 61: 1–12.

Yoo, S.D., Cho, Y.H., and Sheen, J. (2007). Arabidopsis mesophyllprotoplasts: A versatile cell system for transient gene expressionanalysis. Nat. Protoc. 2: 1565–1572.

Zoschke, R., Liere, K., and Börner, T. (2007). From seedling tomature plant: Arabidopsis plastidial genome copy number, RNAaccumulation and transcription are differentially regulated duringleaf development. Plant J. 50: 710–722.

Translation in Arabidopsis Mitochondria 1867

DOI 10.1105/tpc.113.111294; originally published online May 30, 2013; 2013;25;1855-1867Plant Cell

Elwira Sliwinska and Hanna JanskaMalgorzata Kwasniak, Pawel Majewski, Renata Skibior, Aleksandra Adamowicz, Malgorzata Czarna,

MitochondriaArabidopsisTranslation in Gene Encoding Mitochondrial Ribosomal Protein AltersRPS10Silencing of the Nuclear

 This information is current as of March 27, 2021

 

Supplemental Data /content/suppl/2013/05/28/tpc.113.111294.DC2.html /content/suppl/2013/05/21/tpc.113.111294.DC1.html

References /content/25/5/1855.full.html#ref-list-1

This article cites 47 articles, 21 of which can be accessed free at:

Permissions https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists