Mutations in the plant-conserved MTERF9 alter chloroplast gene expression, development and tolerance to abiotic stress in Arabidopsis thaliana Pedro Robles, José Luis Micol and Víctor Quesada* Instituto de Bioingeniería, Universidad Miguel Hernández, Campus de Elche, 03202 Elche, Spain *Corresponding author, e-mail: [email protected] The control of organelle gene expression in plants is far from fully understood. The characterization of mutants in Arabidopsis thaliana is assigning an increasingly prominent role to the mitochondrial transcription termination factors (mTERFs) in this process. To gain insight into the function of mTERF genes in plants, we took a reverse genetics approach to identify and characterize A. thaliana mTERF-defective mutants. Here we report the characterization of the mterf9 mutant, affected in an mTERF protein functionally conserved in plants and targeted to chloroplasts. Loss of MTERF9 results in defective chloroplast development, which is likely to cause paleness, stunted growth and reduced mesophyll cell numbers. Expression analysis of different plastid genes revealed reduced levels of plastid-encoded polymerase (PEP)-dependent transcripts and increased levels of transcripts dependent of nucleus-encoded polymerase. mterf9 plants exhibited altered responses to sugars, ABA, salt and osmotic stresses, and the microarray data analysis showed modifications in MTERF9 expression after salt or mannitol treatments. Our genetic interactions results indicate a functional relationship between MTERF9 and the previously characterized MDA1 gene, and between MDA1 and some plastid ribosomal genes. MDA1 and MTERF9 were up-regulated in the mterf9 and mda1 mutants, respectively. Moreover, 21 out of 50 genes were commonly co-expressed with MDA1 and MTERF9. The analysis of the MDA1 and MTERF9 promoters showed that both were rich in stress-related cisregulatory elements. Ours results highlight the role of the MTERF9 gene in plant biology and deepens in the understanding of the functional relationship of plant mTERF genes. Abbreviations – ABA, abscisic acid; MTERF9; mitochondrial transcription termination factor, mTERF; NEP, Nuclear-encoded polymerase; PEP, plastid-encoded polymerase; PTAC, plastid transcriptionally

active

chromosome;

qRT-PCR,

quantitative

real-time

PCR;

sqRT-PCR,

semiquantitative real-time PCR; T-DNA, transferred DNA; WT, wild-type.

This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/ppl.12307 This article is protected by copyright. All rights reserved

Introduction Plastids and mitochondria are organelles with their own genomes inherited from the ancestral endosymbionts from which they derive. In the last few years, the advances made in the characterization of plastid-transcriptional machinery are revealing that transcription in this organelle is much more complex than initially thought. A small family of genes named RpoT (RNA polymerase T7 phage-type), encoding monomeric RNA polymerases (nuclear-encoded RNA polymerases; NEP), is responsible for the transcription of a number of chloroplasts and/or mitochondria genes in angiosperms and the moss Physcomitrella patens (Hess and Börner 1999, Liere and Börner 2007, Shiina et al. 2005). Unlike mitochondria, the plastid genome harbors information for a second RNA polymerase, similarly to those of prokaryotes [plastid-encoded RNA polymerase (PEP); Allison et al. 1996, Allison 2000, Igloi and Kössel 1992]. Transcription initiation by PEP requires sigma factors encoded by genes that migrated to the nuclear genome. In this way, the nucleus can control PEP transcription initiation in response to metabolic requirements of the whole cell. Multiple transcription start sites have been identified in the plastid genomes of higher plants. However, the cis-regulatory sites for NEP or PEP binding were not found to be upstream of many of these transcription start sites (Wagner and Pfannschmidt 2006). This suggests the participation of additional transcriptional regulatory factors other than NEP and PEP, which would confer the ability to tailor plastid transcription quickly and efficiently in response to a wide range of intra- and extracellular signals. Mitochondrial transcription termination factors (mTERFs), initially identified and characterized in humans and later in other metazoans (Linder et al. 2005), are good candidates to act as regulatory and/or auxiliary transcription factors in plastids. Accordingly, the Arabidopsis thaliana PTAC15 protein, an mTERF, is a component of the TAC (transcriptionally active chromosome or nucleoid) complex that contains more than 50 different proteins (Pfalz et al. 2006). Three other mTERF proteins, BELAYA SMERT/RUGOSA2 (BSM/RUG2; Babiychuk et al. 2011, Quesada et al. 2011), MDA1 (Robles et al. 2012a) and the product of the At2g34620 gene, have been identified from independent nucleoid preparations of A. thaliana seedlings (Huang et al. 2013). Furthermore, 10 mTERFs have been identified in maize leaves as components of the nucleoid-associated proteome (Majeran et al. 2012). Recently, Hammani and Barkan (2014) found that the ZmTERF4 gene, the maize orthologue of A. thaliana BSM/RUG2, promotes the splicing of group II introns in chloroplasts extending mTERF functions to post-transcriptional regulation. The first mTERF identified was the human MTERF1, which was described as a protein that strongly stimulates heavy strand transcription termination in mitochondria based on in vitro data (Kruse et al. 1989). The MTERF1 function in mitochondria has been exhaustively deciphered and, besides transcription termination, it has been shown to participate in initiation and modulation of mitochondrial DNA (mtDNA) replication (Roberti et al. 2009). Nevertheless, recent results from Terzioglu et al. (2013) based on the study of Mterf1 knockout mice do not support a role for MTERF1 in the regulation of mitochondrial rRNA synthesis and translation; rather MTERF1 seems to act This article is protected by copyright. All rights reserved

blocking transcription on the L-strand of mitochondrial DNA, preventing transcription interference at the L-strand promoter. In vertebrates, four mTERF subfamilies have been described (MTERF1-4). Other vertebrate MTERFs have also been characterized: MTERF2 (Wenz et al. 2009), MTERF3 (Park et al. 2007) and MTERF4 (Cámara et al. 2011). Like MTERF1, MTERF2 and MTERF3 bind to the H1 transcription initiation site in mtDNA, as revealed by ChIP assays (Park et al. 2007, Wenz et al. 2009). Transcription initiation is stimulated by MTERF2, but is repressed by MTERF3. Mice Mterf3 and Mterf4 are essential genes because knockouts are lethal (Cámara et al. 2011, Park et al. 2007), whereas inactivation of mice MTERF2 severely reduces the expression of the genes encoding mitochondrial respiratory complexes (Wenz et al. 2009). MTERF4 participates in mitochondrial translation by targeting rRNA methyltransferase NSUN4 to ribosomes (Cámara et al. 2011, Spåhr et al. 2012, Yakubovskaya et al. 2012). The genetic and molecular characterization of mTERF genes has recently been initiated in plants (Robles et al. 2012b). Higher plants genomes encode more mTERFs than animal genomes (Kleine 2012). Thus, more than 30 mTERF genes have been reported in A. thaliana and Oryza sativa, and the vast majority of their protein products act in mitochondria or chloroplasts (Babiychuk et al. 2011, Kleine 2012, Robles et al. 2012a). This expansion of the mTERF family in green plants (especially land plants) might be explained by a functional diversification that would allow mTERFs to carry out functions performed by other classes of proteins in non-plant species (Kleine 2012). Along these lines, mTERFs might differentially regulate the expression of mitochondria and chloroplast genomes in different tissues and/or developmental stages, and in response to environmental stress conditions (Robles et al. 2012b). To study in depth the role of mTERFs in plants, we initiated a functional characterization of the mTERF gene family in A. thaliana by taking a reverse genetics approach (Robles et al. 2012a). Here we describe the characterization of the MTERF9 gene, which is conserved in monocots and dicots. The mterf9 plants exhibited defective chloroplasts, altered expression of chloroplast genes, stunted growth and reduced chlorophyll levels and mesophyll cell number. Furthermore, loss of MTERF9 function alters the response to ABA and different environmental stresses. We found that MTERF9 and the previously characterized mTERF gene MDA1 are functionally related. The genetic interactions observed between mda1, mterf9 and some ribosome-defective mutants suggest a link between plastid ribosomal translation and mTERF activity. This study goes further into the characterization and functional relationship of plant mTERF genes.

Materials and methods Plant material and growth conditions Plant cultures and crosses were performed as described by Ponce et al. (1998) and Berná et al. (1999), respectively. Seeds of the Arabidopsis thaliana (L.) Heynh. wild-type (WT) accession Columbia-0 This article is protected by copyright. All rights reserved

(Col-0) were obtained from the Nottingham A. thaliana Stock Centre (NASC). Seeds of the transferred DNA (T-DNA) insertion lines [N510822, N591486, N595863, N620765, N643391, N651197, N651198, N853938, N857510 and N864365] were provided by the NASC and are described at the SIGnAL web site (Alonso et al. 2003, http://signal.salk.edu). The rug2-2, mda1-1 and mda1-2 mutants in a Col-0 genetic background have been characterized in previous works (Quesada et al. 2011, Robles et al. 2012a). soldat10 [in a Landsberg erecta (Ler) genetic background] and twr-1 (in a Col-0 genetic background) seeds were kindly provided by Klaus Apel (Boyce Thompson Institute for Plant Research, Ithaca, NY) and Renze Heidstra (Utrecht University, Faculty of Science, Department of Biology, Section Molecular Genetics, Padualaan 8, Utrecht), respectively. Pigment extraction and quantification were performed as previously described (Hricová et al. 2006).

In silico analyses Amino acid sequence comparisons and similarity searches were performed using FASTA (Lipman and Pearson 1985) and BLAST (Altschul et al. 1990). The MDA1, MTERF9 and BSM/RUG2 coexpression gene data were retrieved from Genevestigator (Zimmermann et al. 2004). Gene Ontology (GO) annotation terms and the functional categorization of the co-expressed genes were carried out with the GO annotation tool from TAIR (The Arabidopsis Information Resource; Boyle et al. 2004). The MTERF9 orthologues in photosynthetic species were identified using the PLAZA database for comparative plant genomics (Proost et al. 2009, Van Bel et al. 2012). The subcellular localization of the analyzed proteins was predicted with Target P1.1 (Emanuelsson et al. 2007), Protein Prowler (Bodén and Hawkins 2005) and iPSORT (Bannai et al. 2002). The MTERF9 spatial profile and expression responses under salt and osmotic stress were obtained from the Bio-Analytic Resource (BAR, Winter et al. 2007) and the AtGenExpress Visualization Tool (Kilian et al. 2007), respectively. The MTERF9 expression in response to ABA was visualized by TileViz (Zeller et al. 2009). The spatial, developmental and ABA response expression profiles of the rice MTERF9-orthologous gene Os07g39430 were obtained from the RiceXPro database (Sato et al. 2011). Putative promoter sequences, ~2-kb upstream of the translation start site for the analyzed mTERF and control genes, were obtained from and subjected to the prediction of cis-acting regulatory DNA sequences using PlantPAN (Chang et al. 2008) with the transcription factors selected from rice, maize and A. thaliana.

Morphological and ultrastructural analyses Morphological

measurements

were

taken

with

the

ImageJ

(http://rsb.info.nih.gov/ij/docs/menus/file.html) from the pictures taken by

program

a Leica MZ6

stereomicroscope equipped with a Nikon DXM1200 digital camera. A Student´s t-test was applied to the data obtained, with a significance level of 0.01. Confocal imaging was performed as described by Robles et al. (2012a). For transmission electron microscopy, the mutant and WT plant material was harvested at the same time of the day and prepared as described by Hricová et al. (2006). Samples This article is protected by copyright. All rights reserved

were

visualized

under

a

Zeiss

EM10C

transmission

electron

microscope

(Zeiss,

http://www.zeiss.com). Dry weight was measured in the plants oven-dried overnight at 55°C. Flowering time was scored as the number of vegetative leaves and the number of days at bolting of plants grown under continuous light.

Identification of T-DNA in insertional lines DNA was extracted from T3, T4 or T5 mterf9, rpl24, rpl31 and rps5 mutant plants, from F2 and F3 double mutant segregating plants, and from F1 plants derived from an mterf9 x twr-1 cross. DNA was PCR-amplified using primers (RP and LP) designed by the T-DNA Primer Design (http://signal.salk.edu/tdnaprimers.2.html) Tool, which hybridized with the genomic sequences flanking the insertions in combination with T-DNA specific primers (LB1, LBb1.3 or LBp743 primers, Table S1).

RNA extraction and semi-quantitative RT-PCR (sqRT-PCR) Total RNA was extracted using TRIsure (Bioline) and DNase I treated, following the manufacturer’s instructions, from 80 mg of Col-0 and mterf9 of 14 days after stratification (DAS) seedlings. RNA was ethanol-precipitated and resuspended in 40 μl of RNase-free water. Two to four micrograms of each sample were reverse-transcribed using random hexamers/primers and the PCR amplifications of first-strand cDNA were performed as described by Quesada et al. (1999). Then 1 μl of the resulting cDNA solution was used for the sqRT-PCR amplifications. To detect the MTERF9 transcripts in Col0 and mterf9, different primer combinations were used as described in Table S1 and Fig. 2A.

Quantitative RT-PCR (qRT-PCR) Total RNA was extracted from 50–70 mg of 14-day-old seedlings of the WT Col-0 and the mda1-1, mterf9 and twr-1 mutants, as well as from 3-week-old roots and vegetative leaves, 45-day-old stems, cauline leaves and flowers of Col-0. RNA was DNase I-treated following the manufacturer’s instructions. RNA was ethanol-precipitated and resuspended in 40 μl of RNase-free water. Two micrograms of RNA of each sample were reverse-transcribed using random hexamers/primers, as described by Quesada et al. (1999). cDNA preparations and qPCR amplifications were carried out in an ABI PRISM 7000 Sequence Detection System (Applied Biosystems) as described in Robles et al. (2012a). Oligonucleotides (Table S1) were designed as described in Quesada et al. (2011). Each 20-µl reaction mix contained 10 µl of the SYBR-Green/ROX qPCR Master Kit (Fermentas), 0.4 µM of primers and 1 µl of cDNA solution. The relative quantification of the gene expression data was performed using the 2–ΔΔCT method as described in Robles et al. (2012a). Each reaction was performed in three replicates, three different biological replicates were used and the expression levels were normalized to the CT values obtained for the housekeeping OTC gene (Quesada et al. 1999). The

This article is protected by copyright. All rights reserved

results obtained with the OTC gene were always reproducible when other housekeeping genes were used in parallel as internal controls in qRT-PCR experiments.

Germination and growth sensitivity assays For the germination assays, sowings were performed as described in Robles et al. (2012a) on Petri dishes of GM agar medium (Murashige and Skoog medium containing 1% sucrose) supplemented with NaCl (0, 100, 150 and 200 mM), KCl (0, 100, 125 and 150 mM), mannitol (0, 300, 400 and 500 mM), ABA (0, 1, 2, 3 and 6 μM), sucrose (3%, 6% and 10%) or glucose (6%). Seedling establishment, considered as those seedlings exhibiting green and fully expanded cotyledons, was scored 4, 7, 10 and 12 DAS on Petri dishes kept at 20±1°C under 72 μmol m–2 s–1 of continuous light. The NaCl and ABA responses during plant growth were evaluated as described in Robles et al. (2012a). Temperature-sensitivity assays were performed as previously described (Hricová et al. 2006).

Results Identification of the mterf9 mutant In a reverse genetics screen carried out in Arabidopsis thaliana to identify and characterize mTERFdefective mutants (Robles et al. 2012a), we isolated a WiscDsLox T-DNA line (N857510), which was annotated as affected in the At5g55580 gene. N857510 T3 plants displayed a mutant morphological phenotype (Fig. 1), which was inherited with complete penetrance and constant expressivity. T5 mutant individuals were backcrossed twice to the wild-type (WT) ancestor Columbia-0 (Col-0) to remove additional undesired mutations, which allowed us to determine that the morphological phenotype of N857510 is monogenic and recessive (the studied F2 progenies showed a 3:1 WT:mutant segregation ratio; χ2 = 0.13, P=0.73). By PCR, we confirmed the presence of a T-DNA insertion into the fourth intron of the At5g55580 gene (see Materials and Methods), consistent with the SIGnAL website annotation (http://signal.salk.edu). Our PCR analyses showed that all the F2 mutant plants tested and their F3 selfed progeny were homozygous for the T-DNA insertion in the At5g55580 gene. We named this mutant mterf9 according to Kleine classification (Kleine 2012). No morphological or developmental mutant phenotypes were observed in any of the N643391, N651197, N651198, N620765, N853938 and N864365 lines, which putatively harbor different T-DNA insertions upstream the ATG of the coding region of the At5g55580 gene. The effect of the T-DNA insertion on the expression of the At5g55580 (MTERF9) gene in the mterf9 mutant was analyzed. For this purpose, RNA was extracted from 2-week old mutant and WT plants, which was reverse-transcribed and PCR-amplified using different combinations of primers (Fig. 2A; Table S1). The oligonucleotides RP and R2 flanking the T-DNA insertion yielded a PCR product of the expected size when using Col-0 cDNA or genomic DNA, but no amplification was

This article is protected by copyright. All rights reserved

detected from mterf9 cDNA or genomic DNA (Figs. 2A, B). Besides, the use of the F1 and R1 primers, hybridizing upstream of the T-DNA, allowed us to detect MTERF9 transcripts from Col-0 and mterf9 cDNAs (Fig. 2B). When the F1 primer was combined with the T-DNA left border (LB) specific primer LBp745, a chimeric transcript was identified in mterf9, but not in Col-0 (Fig. 2C). Translation of this chimeric transcript is predicted to produce a truncated protein of 391 amino acids, lacking 105 residues of the C-terminal part of the WT protein, including two full-length mTERF motifs (Fig. 2D). Mokry et al. (2011) dubbed twr-2 the N857510 line and found it allelic to the EMS-derived twr-1 mutant which carries a mutation causing a premature stop codon in the At5g55580 gene, that removes the last 30 residues of the C-terminus of its protein product (Fig. 2D). The twr-1 mutant was partially characterized: Mokry et al. (2011) reported meristem defects in the twr-1 plants and that they produced fewer stems and enhanced floral termination resulting in reduced seed set. Under our growth conditions, twr-1 displayed a morphological phenotype similar to that of mterf9, although the twr-1 plants grew more slowly, were smaller, paler and exhibited deeper leaf indentations than mterf9. The F1 progeny of the twr-1 × mterf9 cross exhibited a mutant phenotype to some extent intermediate between those of the parentals (Fig. S1). Besides, all the F2 plants from the selfed F1 individuals showed the Mterf9 phenotype, which was stronger in a quarter of them, as expected for homozygous twr-1/twr-1 plants (χ2=1.30; P=0.24). These results confirmed that twr-1 and mterf9 are allelic, as we corroborated by genotyping (see Materials and Methods).

Plant development is impaired in mterf9 Loss of MTERF9 function pleiotropically affects development. In mterf9 leaves, the lamina was not uniformly pale and the vasculature was distinguishable on the paler green interveinal regions. This phenotype is, to some extent, similar to that of the reticulated or variegated mutants (González-Bayón et al. 2006, Quesada et al. 2011, Fig. 1C–F). In addition, mterf9 leaves were markedly dentate (Fig. 1E, F). Our confocal microscopy observations of the internal leaf structure of mterf9 revealed both reduced mesophyll cell density and chloroplasts size compared with those of the wild-type (Fig. 1G, H). To further explore the effect of defective MTERF9 on chloroplast development, we studied chloroplast ultrastructure in mterf9 and Col-0 leaf mesophyll cells by transmission electron microscopy. Consistently with the confocal microscopy results, mterf9 chloroplasts were found smaller than those of Col-0 and their internal structure was altered (Fig. 3A–F). Accordingly, thylakoid lamellae were enlarged and fewer starch grains were found in mterf9. We detected no vacuole or plastoglobuli accumulation (Fig. 3C–F). Chloroplast number was similar in Col-0 and mterf9. Mitochondrial morphology was also similar in mterf9 and Col-0 (Fig. 3G, H). As expected from the chloroplast defects found in mterf9 cells, the mutant plants were pale, in particular at early stages of development because germinated seedlings are bleached (Fig. 1A, B). Accordingly, chlorophyll a and b and carotenoid levels showed a significant reduction in mterf9 when This article is protected by copyright. All rights reserved

compared with Col-0 (Table S2). Moreover, mterf9 plants exhibited hindered growth and a reduction in size. Accordingly, fresh and dry weights, as well as the length of the main root and stem were reduced in mterf9 compared with the wild type (Table S3). Besides, we found that mterf9 plants flowered earlier than the WT, a phenotypic trait also shared by other mTERF-deficient mutants, such as mda1 (Robles et al. 2012a), rug2-1 (Quesada et al. 2011) or soldat10 (Meskauskiene et al. 2009) (Table S4). The observed perturbation in chloroplast development and the pale pigmentation and small size of the mterf9 plants led us to study whether photoautotrophic growth was altered in the mutant. Accordingly, we sowed WT and mterf9 seeds in media supplemented or not with 1% sucrose (see Materials and Methods). Absence of sucrose did not appreciably affect either the growth rate or morphology of mterf9 plants [4.7±1.2% and 3.7±2.5% of the WT and mutant seedlings, respectively, showed arrested development in the absence of sucrose 21 days after stratification (das)]. In contrast, mterf9 paleness was sensitive to sucrose concentrations above 1%. In addition, high sucrose (6%) made mterf9 to accumulate anthocyanins, and they were unable to fully expand cotyledons as scored 5 das (Fig. S2A, B). Consistently with this, mterf9 seeds led to lower percentages of seedling establishment (the ability to form fully expanded green cotyledons) than Col-0 on media supplemented with 3%, 6% and 10% sucrose (Fig. S2C) or 6% glucose (43.1±5.2% and 22.1±12.05% of Col-0 and mterf9, respectively, 14 das).

Ascertaining MTERF9 putative functions through in silico analysis MTERF9 is a protein of 496 amino acids and 57.16 kDa that contains 6 mTERF motifs, as predicted by SMART (Fig. 2D; http://smart.embl-heidelberg.de/). According to the GFP-fusion results from Babiychuk

et

al.

(2011)

and

our

own

bioinformatic

(http://www.cbs.dtu.dk/services/TargetP/),

predictions Protein

using

Target

P1.1

Prowler

(http://pprowler.itee.uq.edu.au/pprowler_webapp_1-2/) and iPSORT (http://ipsort.hgc.jp/), MTERF9 is a chloroplast protein. In line with this, MTERF9 has been recently included as a component of the A. thaliana reference plastid proteome, according to Huang et al. (2013). BLAST homology searches revealed that the MTERF9 closest paralog was the mTERF protein encoded by the At4g38160 gene (29.1% identity and 64.8% similarity).We investigated MTERF9 conservation in different photosynthetic organisms by using the PLAZA database (http://bioinformatics.psb.ugent.be/plaza/), a comparative plant genomics resource. We identified 24 putative orthologues in green algae, mosses, lycopsids, monocotyledonous and dicotyledonous species (Fig. S3). TargetP1.1 predicted most of them to be chloroplastic, including the product of the rice orthologous gene (Os07g39430; cTP=0.862; Fig. S3B). Besides, the ortholog protein from maize, ZmTERF9, was identified in the nucleoids of plastid leaves (Majeran et al. 2012). In comparison to MTERF9, the ZmTERF9 and the Os07g39430-encoded protein had a similar number of mTERF motifs (7) and residues (503 and 508,

This article is protected by copyright. All rights reserved

respectively), and their level of homology with MTERF9 (54.5% and 48.4% identity, respectively) was higher than that of MTERF9 with any A. thaliana paralog. In accordance with the PLAZA database, Arabidopsis lyrata AL8G20980 and Populus trichocarpa PT01G36520 proteins showed the highest levels of identity with MTERF9 (93.8% and 68.1%, respectively; Fig. S3A). Our phylogenetic analysis revealed that monocotyledonous, dicotyledonous and lower-plant MTERF9 orthologues were grouped into different clades (Fig. S3B). Apart from plants, MTERF9 displayed the highest homology with the members of the MTERF3 sub-family from metazoans (e.g. 24.4% identity and 56.5% similarity with human MTERF3; 21.1% identity and 54.7% similarity with Drosophila melanogaster mTERF3). A sequenced-based alignment of A. thaliana MTERF9 and human MTERF3 proteins revealed the conservation of several hydrophobic residues within mTERF-motifs 1 to 4, described to form hydrophobic interactions required to stabilize the mTERF-motifs, as well as some conservation of proline residues (e.g. in mTERF-motifs 3, 4 and 6) involved in the right twist of the mTERF repeats (Spåhr et al. 2010; Fig. S4). To shed light onto the function of MTERF9, we used the Co-Expression tool from Genevestigator (https://www.genevestigator.com/gv/plant.jsp; Zimmermann et al. 2004) to find genes co-expressed (candidates to be co-regulated) with MTERF9. We extended the analysis to the mTERF genes MDA1 and BSM/RUG2 given the similarity between both the phenotypes of mterf9, mda1 and rug2-2 and the functional relationship between MDA1 and MTERF9 (see later). A list of the top 50 most correlated genes was obtained for each target gene using the dataset of anatomical parts. The Pearson’s correlation coefficient (PCC) values obtained ranged from 0.73 to 0.94 (Table S5). MDA1 and MTERF9 were co-expressed (PCC = 0.82; Table S5). We analyzed the genes co-expressed with MDA1 and MTERF9 by using the Gene Ontology (GO) Annotation tool from TAIR (http://www.arabidopsis.org/tools/bulk/go/index.jsp). As regards the GO terms for “cellular components”, the main categories for the MTERF9 co-expressed genes were “chloroplast” and “plastid”, associated with 37 and 12 genes, respectively, representing in both cases almost a 5-fold enrichment compared to the whole A. thaliana genome (Table S6). Regarding the GO terms for “biological processes”, the two most represented categories among the MTERF9 co-expressed genes were “other cellular processes” and “other metabolic processes”. Similar results were obtained for MDA1 (Table S6). Twenty-one genes (43.7%) were commonly co-expressed with MDA1 and MTERF9 (Table S5), most of them (71.4%) included in the cellular component category of “chloroplast” (Table S6) such as At2g36000, At4g20130, At3g20230 and At2g33430, which encode an mTERF, the PTAC14 protein, the L18p/L5e ribosomal protein and the MULTIPLE ORGANELLAR RNA EDITING FACTOR 2 (MORF2), respectively, all involved in plastid gene expression (Table S5). Regarding RUG2, 13 (24%) and 9 (18%) of the top 50 co-expressed genes were common to those of MDA1 and MTERF9, respectively (Table S5).

This article is protected by copyright. All rights reserved

MTERF9 expression analysis We investigated the spatial expression pattern of the MTERF9 gene using qRT-PCR and RNA extracted from different organs. MTERF9 transcripts were detected in all the organs studied. In comparison to roots, MTERF9 was expressed at higher levels in buds, open flowers, stems and rosette leaves [up-regulated 4.6-fold (4.6±1.1; P

Mutations in the plant-conserved MTERF9 alter chloroplast gene expression, development and tolerance to abiotic stress in Arabidopsis thaliana.

The control of organelle gene expression in plants is far from fully understood. The characterization of mutants in Arabidopsis thaliana is assigning ...
458KB Sizes 0 Downloads 10 Views