Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

ECD1 functions as an RNA-editing trans-factor of rps14-149 in plastids and is required for early chloroplast development in seedlings

ECD1 functions as an RNA-editing trans-factor of rps14-149 in plastids and is required for early... Chloroplast development is a highly complex process and the regulatory mechanisms have not yet been fully charac- terized. In this study, we identified Early Chloroplast Development 1 (ECD1), a chloroplast-localized pentatricopeptide repeat protein (PPR) belonging to the PLS subfamily. Inactivation of ECD1 in Arabidopsis led to embryo lethality, and abnormal embryogenesis occurred in ecd1/+ heterozygous plants. A decrease in ECD1 expression induced by RNAi resulted in seedlings with albino cotyledons but normal true leaves. The aberrant morphology and under-developed thylakoid membrane system in cotyledons of RNAi seedlings suggests a role of ECD1 specifically in chloroplast development in seedlings. In cotyledons of ECD1-RNAi plants, RNA-editing of rps14-149 (encoding ribosomal protein S14) was seriously impaired. In addition, dramatically decreased plastid-encoded RNA polymerase-dependent gene expression and abnormal chloroplast rRNA processing were also observed. Taken together, our results indicate that ECD1 is indispensable for chloroplast development at the seedling stage in Arabidopsis. Keywords: Arabidopsis, chloroplast development, cotyledon, Early Chloroplast Development 1 (ECD1), early stages, pentatricopeptide repeat protein (PPR), RNA editing. Introduction Chloroplasts are not only the exclusive organelles that perform polymerases: plastid-encoded bacterial-type RNA polymer- photosynthesis but they are also responsible for many other ase (PEP) and nuclear-encoded phage-type RNA polymerase biosynthetic processes, such as the synthesis of amino acids, (NEP) (Hedtke et  al., 1997; Liere and Maliga, 2001; Börner hormones, and metabolites (Sakamoto et al., 2008). The devel- et al., 2015). Given their increasing activity during chloroplast opment of functional chloroplasts is a prerequisite for photo- development, PEPs are clearly crucial for chloroplast develop- synthesis and is also tightly co-ordinated with plant growth and ment at early stages of plant growth (Mullet, 1993). During development. Chloroplast gene expression is crucial for chlo- chloroplast biogenesis, plastid ribosomal proteins are required roplast development, and is carried out by two kinds of RNA to establish a functional chloroplast translational apparatus and © The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Experimental Biology. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 3038 | Jiang et al. deficiency of these proteins leads to lethality (Tiller and Bock, and which belongs to the PLS subgroup of the PPR family. 2014). The successful assembly of ribosomal proteins in chloro- Disruption of ECD1 leads to embryo lethality and RNAi lines plast is therefore very important for chloroplast development. display albino cotyledons but normal leaves. Aberrant chloro- Derived through endosymbiosis from cyanobacteria, chlo- plast ultrastructure and deficient RNA-editing of rps14-149 roplasts are semi-autonomous organelles that have their own (encoding ribosomal protein S14) in plastids were detected in genome. However, there are only about 100–150 genes in the cotyledons of the ECD1-RNAi transgenic plants. Mutation of plastid genome, the products of which are mainly involved in ECD1 also resulted in decreased expression of PEP-dependent photosynthesis and plastid gene expression (Sato et al., 1999). genes and abnormal rRNA processing. Our results indicate The vast majority of chloroplast proteins (>2000) are encoded that ECD1 plays a vital role in chloroplast development in in the nucleus, translated in the cytosol, and then imported seedlings. into the chloroplast. Thus, the formation of functional chlo- roplasts relies on co-ordination of gene expression between Materials and methods the plastid and the nucleus. Chloroplast gene expression is regulated by a set of nuclear-encoded factors. Among these, Plant materials and growth conditions pentatricopeptide repeat (PPR) proteins, which constitute one We obtained the T-DNA insertion line CS16045 (ecotype Columbia) of the largest protein family in land plants, have been dem- from the Arabidopsis Biological Resource Center (https://abrc.osu.edu/; onstrated to play important roles in chloroplast gene expres- last accessed 27 Apr il 2018). This line is the same as the one used in previous sion and function. Mutations of PPR genes usually result in studies by Tzafrir et al. (2004) and Cushing et al. (2005). The T -DNA inser- tion was confirmed by PCR with T-DNA-specific primers. Seeds of wild- seedling-lethal or embryo-lethal phenotypes. The Arabidopsis type and mutant plants were surface-sterilized after incubation at 4 °C for genome encodes more than 450 members of this family (Lurin 3 d to synchronize germination, then sown on Murashige and Skoog (MS) et  al., 2004; Shikanai and Fujii, 2013), and almost all of them medium containing 2% (w/v) sucrose. Plants were grown in soil under a –2 –1 are predicated to localize to plastids or mitochondria (Lurin 12/12-h light/dark cycle with a photon flux density of 120 μmol m s et  al., 2004). Members of the family are characterized by the at 22  °C. For lincomycin and spectinomycin treatment, the wild-type seeds were surface-sterilized and placed on MS media with the addition PPR motif, which appear as tandem repeats of a highly degen- –1 of 500 μM lincomycin or 50 mg l spectinomycin. The seedlings were erate unit of 35 amino acids (Small and Peeters, 2000; Lurin harvested after 7 d. et  al., 2004). The PPR protein family is classified into P and To produce ECD1 knock-down plants, an RNAi construct for ECD1 PLS subfamilies (Lurin et al., 2004), the latter being specific to was generated. A fragment of 434 bp of the ECD1 gene (from nucleotides land plants. The P subfamily usually does not contain any other 286 to 719) was amplified and inserted into the PFGC5941 vector. The forward restriction endonucleases were NcoI and SwaI, and the reverse conserved motifs except for the canonical PPR (P) motifs. By enzymes were XbaI and BamHI. The constructs were transformed into contrast, the PLS subfamily contains long (L) and short (S) Agrobacterium tumefaciens strain GV3101 and introduced into the wild- PPR-like motifs as well as classic PPR motifs. In addition, type plants by the floral dip method (Clough and Bent, 1998). Transgenic –1 based on the presence of different C-terminal motifs, the PLS plants were selected on MS medium containing 50 μg ml Basta. subfamily is further divided into the PLS, E, E+, and DYW subgroups (Schmitz-Linneweber and Small, 2008). Subcellular localization of GFP proteins PPR proteins have been reported to be involved in almost DNA encoding the 218 N-terminal amino acids of ECD1 was ampli- all stages of chloroplast gene expression. For example, PPR10 fied and ligated into the green fluorescent protein (GFP) fusion vector is required for the accumulation of processed RNAs with the pUC18-35S-sGFP with GFP as a reporter. The controls with mitochon- 5´ or 3´ terminus in the atpI-atpH or psaf-rpl33 intercistronic drial-, chloroplast-, and nuclear-localization signals were FROSTBITE1 region (Pfalz et al., 2009; Barkan and Small, 2014). CRR2 was (FRO1), ribulose bisphosphate carboxylase small subunit (RbcS), and the first reported DYW-PPR protein and it is involved in the PTM-N (Sun et  al., 2011), respectively. The resulting fusion constructs and the control vectors were introduced into Arabidopsis mesophyll pro- intergenic RNA cleavage between rps7 and ndhB (Hashimoto toplasts according to the PEG-mediated method (Kovtun et  al., 2000). et  al., 2003; Shikanai and Fujii, 2013). There are ppr mutants Fluorescence analysis was performed on an LSM 510 Meta confocal laser that affect PEP-dependent gene expression, such as dg1 (Chi scanning system (LSM510; Carl Zeiss, Jena, Germany). et al., 2008). SOT1, a PPR protein with a small MutS-related (SMR) domain has endonuclease activity. Its PPR domain spe- cifically recognizes a 13-nucleotide RNA sequence in the 5´ RNA gel blotting, RT-PCR, and quantitative RT-PCR end of the chloroplast 23S-4.5S rRNA precursor (Zhou et al., Total leaf RNA was extracted from 7-d-old seedlings, and from 14-d-old cotyledons and true leaves using an RNeasy Plant Mini kit (Qiagen). 2017). A r ice mutant, wsl, in which a PPR protein WSL is miss- RNA concentration was determined using thermo NanoDrop 2000. ing, exhibits reduced translation efficiency caused by abnormal Total RNA from seedlings of the wild-type and the ECD1-RNAi-1 splicing of the rpl2 gene (Wang et al., 2017). The PPR protein line was separated on 1.3% (w/v) agarose-formaldehyde gels, blotted to a EMB2654 is required for the trans-splicing of the plastid rps12 nylon membrane, and subsequently hybridized with a probe labeled with transcript and its binding site is localized on one of the intron P. The probes were prepared by PCR amplification and labeled using the Prime-a-Gene Labeling System (SGMB01-Promega-U1100). halves (Aryamanesh et al., 2017). RNA was used to generate first-strand cDNA in a 20-μl reaction using Despite the presence of numerous PPR proteins in higher the Superscript III cDNA synthesis system (Invitrogen). The resulting plants, their functions in the regulation of chloroplast devel- cDNA samples were used as templates for RT-PCR analysis. Quantitative opment has not yet been elucidated. Here, we report a novel RT-PCR was performed using the SYBR Premix ExTaq Kit (Takara) chloroplast factor, Early Chloroplast Development 1 (ECD1), following the manufacturer’s instructions with a Light Cycler 480 system. The expression level was normalized to that of an ACTIN control. that is involved in early chloroplast development in seedlings Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 RNA-editing factor ECD1 and chloroplast development | 3039 Histochemical GUS staining Yeast two-hybrid assays Tissues were incubated in cold 90% (v/v) acetone and placed in a staining The CDS of ECD1 without the first 150 bp encoding the transit pep- buffer (100 mM sodium phosphate buffer, pH 7.2, 0.2% Triton X-100, tide was amplified by PCR and cloned into pGBKT7 DNA-BD as bait. 10 mM potassium ferrocyanide, 10 mM potassium ferricyanide, 0.25M Sequences encoding the mature MORF2, MORF3, MORF6, MORF8, EDTA, and 1  mM X-gluc). After vacuum-infiltration for 15–30  min, and MORF9 proteins were cloned into pGADT7 as prey. The prey samples were incubated for 16–24 h at 37 °C. β-Glucuronidase (GUS)- and bait constructs were co-transformed into Y2HGold yeast cells. The stained tissues were cleared with an ethanol series of 20% (v/v), 30% transformation was performed using the Matchmaker Gold Yeast Two- (v/v), and 50% (v/v) for 30 min in turn, and finally incubated in a solu- Hybrid System (Clontech) according to the manufacturer’s instructions. tion of 70% (v/v) ethanol and 30% (v/v) acetic acid for at least 30 min Interactions were determined by growing diploid yeast colonies on syn- –1 until the tissues became transparent enough to observe under a dissecting thetic dropout (SD) medium containing 40 μg ml X-α-Gal (5-bromo- Olympus SZX16 microscope. 4-chloro-3-indolyl-α-d-galactopyranoside) without tryptophan, leucine, and histidine. Protein isolation and immunoblot analysis RNA immunoprecipitation assays Total proteins were prepared as previously described (Martínez-García et al., 1999). Protein concentrations were determined using the Bio-Rad RIP assays were performed as described previously (Kim et  al., 2012) DC protein assay. For immunoblot analysis, total proteins were separated using 7-d-old 35S::ECD1-FLAG transgenic seedlings. Anti-FLAG by SDS-PAGE and transferred to nitrocellulose membranes. The mem- M2 magnetic beads were obtained from Sigma (M8823). Protein A/G branes were incubated with specific primary antibodies, and the signals Sepharose incubated with pre-immune serum was used as the control. were detected using a Pro-Light HRP Chemiluminescent Kit (Tiangen RNA was isolated by phenol-chloroform isoamyl alcohol extraction and Biotech). PsaA, D1, LHCII, Cytb6, Cytf, CF0II, and RPS14 were then analysed by qRT-PCR. expressed and purified in-house at our laboratory and used to generate polyclonal antibodies in rabbits. The antisera for RPL2 and RPS3 were Electrophoretic mobility shift assays provided by Tiegang Lu, and the anti-FLAG antibody was obtained from Abmart (www.ab-mart.com/; last accessed 27 April 2018). EMSAs were carried out using a LightShift Chemiluminescent RNA EMSA Kit (Thermo 20158). After incubation for 20 min at 25 °C, the samples were resolved on a 6% Tris-borate gel in 0.5× TBE buffer, trans- Analysis of RNA editing ferred to a nylon membrane, and subsequently processed using a chemi- A series of specific primers were used to amplify the regions of the genes luminescent detection kit (Thermo 89880). The 5´-end biotin-labeled containing the editing sites in Arabidopsis (Cai et  al., 2009) from the oligoribonucleotides rps14-80 and rps14-149 were synthesized and cDNA using RT-PCR, and the products were sequenced directly (for a list labeled by Takara Bio Inc. To produce recombinant MBP-ECD1 pro- of primers used in this study see Supplementary Table S1 at JXB online). teins, the coding sequence of ECD1 lacking the transit peptide sequence The levels of RNA editing were estimated by the relative heights of the was PCR-amplified, digested with SacI and NotI, and inserted into pET - peaks of the nucleotide in the sequence analysed. Plasmids prepared from MALc-H (Pryor and Leiting, 1997). Recombinant protein was expressed approximately 90 independent colonies of each sample were sequenced to and then purified by amylose affinity chromatography according to the determine the RNA-editing efficiency of rps14-80 and rps14-149. manufacturer’s instructions (New England BioLabs). Transmission electron microscopy Results For TEM processing, wild-type and ECD1-RNAi-1 leaves from 7-d-old plants, and from cotyledons and true leaves from 14-d-old plants were Mutations in ECD1 produce defects in embryogenesis collected. The tissue was cut into small pieces and fixed in 3% glutaral- dehyde in phosphate buffer for 4 h at 4 °C. After fixation, the tissue was To study the detailed mechanisms of chloroplast develop- rinsed in phosphate buffer 3–4 times and then post-fixed in 1% OsO ment, we obtained a series of T-DNA insertion lines from the overnight at 4  °C. After rinsing in phosphate buffer again, the samples Arabidopsis Biological Resource Center, the products of which were dehydrated in an ethanol series, infiltrated with a graded series of are predicted to be candidates for chloroplast biogenesis fac- epoxy resin in epoxy propane, and embedded in Epon 812 resin. Thin sections were obtained using a diamond knife on a Reichert OM2 ultra- tors. An embryo-lethal line CS16045 of the gene AT3G49170, microtome, stained with 2% uranylacetate, pH5.0, followed by 10  mM designated as ecd1 (originally called emb2261), attracted our lead citrate, pH12, and viewed with a transmission electron microscope attention for further investigation. Failure to identify any prog- (JEM-1230; JEOL). eny homozygous for the ecd1 mutant allele suggested that the mutation causes embryonic lethality. We dissected the develop- Analysis of embryo development ing siliques and assessed the seeds under a dissecting micro- Embryos were excised from wild-type and ecd1+/ siliques at different scope. In wild-type siliques, all the ovules developed normally, developmental stages and cleared in Hoyer’s solution (7.5 g gum arabic, while in the heterozygous ecd1/+ siliques, some ovules were 100 g chloral hydrate, and 5 ml glycerol in 30 ml water) as described by white (Fig. 1A). In 33 siliques from the heterozygous ecd1/+ Meinke (1994). Embryo development was studied microscopically using plants, 177 out of 726 ovules were white, making the ratio an Olympus BH-2 microscope equipped with Nomarski optics. of white to green ovules 1:3 (χ =0.166, P>0.05) (data not shown). In older siliques, the white ovules became shrunken Bimolecular fluorescence complementation assays and aborted. BiFC assays were performed as previously described (Walter et  al., 2004). To deter mine precisely the stage of embryogenesis dur ing Full-length cDNA of ECD1 was cloned into pSAT4A-nEYFP-N1, and full- which the ECD1 mutant arrested development, develop- length cDNAs of MORF2 and MORF9 (multiple organellar RNA-editing ing seeds at various stages from self-pollinated heterozygous factor) were cloned into pSAT4A-cEYFP-N1. The plasmids were co-trans- plants were cleared and observed using differential interfer- formed into protoplasts. Yellow fluorescent protein (YFP) was imaged using a confocal laser scanning microscope (LSM510; Carl Zeiss, Jena, Germany). ence contrast microscopy. The normal wild-type embryos Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 3040 | Jiang et al. underwent typical developmental stages, ranging from pre- The ECD1 gene encodes a chloroplast PPR protein globular, globular, heart-shaped, torpedo-shaped, through belonging to the PLS subfamily to cotyledon and maturity. However, in the heterozygous The ECD1 gene encodes a putative protein of 850 amino acids ecd1/+ siliques, although the homozygous mutant embryos with a predicated molecular mass of 95.5 kDa. The N-terminal consistently initiated cotyledons and showed continued 50 amino acids are predicated by ChloroP1.1 (http://www.cbs. growth and cell division, beyond the heart stage the devel- dtu.dk/services/ChloroP/; last accessed 27 April 2018) to con- opment of the embryos was significantly slower than that stitute a chloroplast transit peptide. Sequence analysis revealed of the wild-type (Fig.  1B). Mutant embryos consistently that the ECD1 protein contains 17 PPR or PPR-like (P, L, failed to elongate, developing instead as v-shaped embryos and S) motifs, together with one E motif, one E+ motif, and with wide, stunted cotyledons and no hypocotyl (see also one DYW motif in the C-terminal part (see Supplementary Cushing et al., 2005). Fig. S1A). It belongs to the PLS subgroup of the PPR protein family. Protein alignment showed that ECD1 shares significant identity at the amino acid level with proteins from Brassica, Knock-down of ECD1 results in a cotyledon-specific grape (Vitis), eggplant (Solanum), Zea mays, and rice (Oryza) albino phenotype (Supplementary Fig. S1B). To determine the subcellular localization of the ECD1 pro- To further investigate the function of ECD1, we constructed tein, the 218 N-terminal amino acids were fused to the N RNAi lines. A  total of 42 out of 76 RNAi-ECD1 trans- terminus of synthetic GFP (sGFP). The ECD1-GFP fusion genic lines with the abnormal cotyledon phenotype were protein was transiently expressed in Arabidopsis protoplasts obtained (data not shown). In further studies, three RNAi under the control of the cauliflower mosaic virus 35S pro- transgenic lines with a range of stable phenotypes with moter. We observed that the GFP fluorescence merged with respect to white cotyledons and stunted plant growth were the chlorophyll autofluorescence (Fig.  3A), indicating that selected (Fig. 2A, B); however, these lines all had true leaves ECD1 is a chloroplast protein. When ECD1-GFP was tran- that were normal green. RT-PCR showed that the pheno- siently co-expressed with red fluorescent protein (RFP) fused types of these RNAi lines correlated with the expression with pTAC5 (a well-characterized protein known to localize levels of the ECD1 gene (Fig.  2C). These results indicated in nucleoids; Chi et al., 2014), the green and red fluorescence that disruption of the ECD1 gene led to abnormal cotyle- signals within the chloroplasts were found to merge, indicat- dons. The most severely affected line, ECD1-RNAi-1, in ing that ECD1 and pTAC5 were co-localized in chloroplast which the cotyledons were albino, was selected for further nucleoids (Fig.  3B). To further determine the localization of analysis. Fig. 1. Embryogenesis of wild type and ecd1/+ embryos. (A) A heterozygous ecd1/+ mutant silique showing that approximately one-quarter of the ovules are albino compared with the wild-type (WT). Scare bars are 0.5 mm. (B) I–VI, normal embryos of the wild-type: I, pre-globular; II, globular; III, heart-shaped; IV, torpedo-shaped; V, cotyledon; VI, mature. VII–XII, embryos of ecd1/+. VII–IX, embryos are similar to the wild-type; X–XII, development is arrested, with v-shaped embryos with wide, stunted cotyledons and no hypocotyl. Scare bars are 20 μm. (This figure is available in colour at JXB online.) Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 RNA-editing factor ECD1 and chloroplast development | 3041 Fig. 2. Characterization of the ECD1-RNAi transgenic plants. (A) Identification and isolation of RNAi lines with different degrees of inhibition of ECD1 expression (WT: wild-type, ecotype Columbia). Plants were grown on MS medium with 2% (w/v) sucrose for 7 d. Scare bars are 1 mm. (B) Albino cotyledon phenotype of ECD1-RNAi-1 compared with the WT. Plants were grown on MS medium with 2% (w/v) sucrose (3–14 d) or soil (28 d). Scare bars are 1 mm for 3–14 d; 1 cm for 28 d. (C) Reverse transcription PCR (RT-PCR) using specific primers for AT3G49170 or Actin12 for 27 cycles for WT and RNAi lines with different degrees of inhibition of ECD1. (This figure is available in colour at JXB online.) ECD1, intact chloroplasts of overexpressing ECD1-Flag trans- In addition, GUS activity was also detected in rosette leaves, genic plants were isolated and fractionated, and the proteins flower buds, flowers, and siliques, with minimal expression were separated by SDS-PAGE followed by immunoblot anal- also observed in roots (Supplementary Fig. S2). In the flowers, ysis using an anti-FLAG antibody. The ECD1 protein was GUS staining was observed exclusively in green tissues, such as sepals, stamens, and carpels, but not in petals. Taken together, detected in both the stromal and thylakoid fractions (Fig. 3C). these findings showed that ECD1 was widely expressed throughout the plant, but the highest expression was in the Gene expression pattern of ECD1 cotyledons, which was consistent with the albino cotyledon To investigate the expression pattern of the ECD1 gene in phenotype of the ECD1-RNAi transgenic plants. The results Arabidopsis, we made transgenic plants expressing the GUS also indicated that ECD1 expression is developmentally con- protein under the control of the ECD1. The highest expres- trolled and corresponds to early chloroplast development in sion levels were observed in the cotyledons of seedlings. seedlings. Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 3042 | Jiang et al. Fig. 3. ECD1 is localized in the chloroplasts. (A) Subcellular localization of ECD1-GFP. The fluorescence of the ECD1-GFP fusion protein in protoplasts was observed using confocal laser scanning microscopy. Green fluorescence signals, chlorophyll red autofluorescence signals, and merged images are shown. The controls used for transformations are indicated to the right: Mit-GFP, control with mitochondrial localization signal of FROSTBITE1 (FRO1); Chl-GFP, control with the transit peptide of the Rubisco small subunit (RbcS); Nuc-GFP, control with nuclear localization signal of PTM-N (Sun et al., 2011). The scare bar is 10 μm. (B) Co-localization of ECD1-GFP with the pTAC5-RFP protein in chloroplast nucleoids. The red fluorescence of chlorophyll has been adjusted for better contrast. The scare bar is 10 μm. (C) Immunoblot analysis of the ECD1-FLAG fusion protein in chloroplast subfractions. ECD1 localizes to both the stroma and thylakoid membrane fractions. Intact chloroplasts were isolated from 35S::ECD1-FLAG transgenic seedlings and then separated into fractions. Polyclonal antisera were used to detect the ECD1-FLAG fusion protein, the light-harvesting complex II (LHCII), and the Rubisco large subunit (rbcL). (This figure is available in colour at JXB online.) in seedlings. We next examined chloroplast morphology and The ECD1 mutation affects early chloroplast ultrastructure in the transgenic plants using TEM. The wild-type development in seedlings chloroplast ultrastructure was similar in cotyledons and true Together with the chloroplast localization of the ECD1 protein, leaves, with well-developed thylakoid membranes composed of the albino cotyledons of ECD1-RNAi-1 plants suggested that grana connected by stroma lamellae (Supplementary Fig.  S3). the function of ECD1 is related to early chloroplast development Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 RNA-editing factor ECD1 and chloroplast development | 3043 The chloroplasts of albino cotyledons of ECD1-RNAi-1 had of cotyledons and true leaves in the wild-type and transgenic no organized thylakoid membrane system but instead contained plants; however, we did not include the nine sites recently a large number of round or oblong membrane-bound internal found by Ruwe et al. (2013) because of their extremely low vesicles. However, the chloroplasts of true leaves from ECD1- editing efficiency (<10%) even in the wild-type. The results RNAi-1 contained organized thylakoid membranes similar showed that the editing efficiency of the ribosomal protein to those of the wild type. The results indicated that ECD1 is rps14-149 decreased to extreme low levels in the cotyledons involved in chloroplast development in seedlings. of ECD1-RNAi-1 compared to that of the wild-type (Fig. 5). Multiple other sites were also affected to varying degrees (Supplementary Fig.  S4), but no significant differences in The ECD1 mutation affects the accumulation of the editing efficiency were detected in true leaves between proteins of photosynthetic complexes ECD1-RNAi-1 and the wild-type (Supplementary Fig. S6). Since the chloroplast ultrastructure was affected in the ECD1- To test whether the editing deficiency was indirectly caused RNAi transgenic plants, we investigated whether the accumu- by the albino cotyledon phenotype, we pharmacologically lation of proteins of the photosynthetic complexes was different induced albinism by using the plastid translation inhibitor in these plants. Immunoblotting was performed to analyse the lincomycin. Lincomycin treatment results in a severe albino levels of the proteins of each of the thylakoid protein com- phenotype and it has been reported to have severe effects on plexes. The protein levels of PSI (PsaA), PSII (D1 and LHCII), RNA editing (Tseng et al., 2013). Our results showed that in Cytb6/f (Cytb6 and Cytf), and ATPase (CF0II) of cotyledons lincomycin-treated seedlings, the editing efficiency of many in ECD1-RNAi-1 were dramatically decreased compared to sites was significantly reduced or even completely abolished, those of the wild type (Fig. 4). However, the accumulation of including accD-794, accD-58642, petL-5, ndhB-836, ndhD-878, these photosynthetic proteins in true leaves of ECD1-RNAi-1 and ndhF-290 and these were also affected in ECD1-RNAi-1 was similar to that in the wild-type. (Supplementary Fig. S4). However, no obvious editing defi- ciency of rps14-149 was detected in lincomycin-treated seed- lings (Supplementary Figs S4, S5). To rule out the possibility ECD1 is required for RNA-editing of rps14-149 in that the editing defects were unique to lincomycin, we eval- plastids of cotyledons uated the effect of another inhibitor of chloroplast transla- Previous studies have shown that the critical function of the tion, spectinomycin, on editing and obtained similar results DYW domain is involved in RNA editing. To test whether (Figs  S4, S5). We therefore concluded that the mutation of the RNA-editing status in the ECD1-RNAi transgenic plants ECD1 specifically affected RNA editing of rps14-149 in plas- was altered, we sequenced the 34 known editing sites in plastids tids of cotyledons. Fig. 4. Immunoblot analysis of photosynthetic proteins. Total protein was separated by 10% Tricine/SDS-PAGE, electro-blotted, and probed using specific anti-PsaA, anti-D1, anti-LHCII, anti-Cytb6, anti-Cytf, and anti-CF0II antibodies. C and T refer to the proteins in cotyledons and true leaves, respectively, of ECD1-RNA-1 and wild-type (WT) seedlings. Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 3044 | Jiang et al. Fig. 5. Analysis of RNA-editing of rps14 transcripts from wild-type (WT) and ECD1-RNAi-1 seedlings. (A) RT-PCR products containing the rps14-80 and rps14-149 editing sites were directly sequenced. The editing sites of rps14-80 and rps14-149 are indicated by asterisks above the corresponding peaks. (B) The editing efficiency was determined by analysis of approximately 90 independent clones from RT-PCR products for rps14-80 and rps14-149 in the wild-type and ECD1-RNAi-1 line. (This figure is available in colour at JXB online.) Yeast two-hybrid screening with ECD1 as bait identified EMSA (Fig. 7A). As shown in Fig. 7B, binding of MBP-ECD1 several members of the family of multiple organellar RNA- to the rps14-149 oligonucleotide was increased with increasing editing factors (MORF) including MORF2, MORF3, concentrations of the MBP-ECD1 protein. The specificity of MORF6, MORF8, and MORF9 (Fig. 6A). We also confirmed binding was confirmed using the same unlabeled oligoribo- the interactions between ECD1 and the chloroplast-targeted nucleotide as a competitor (Fig.  7C), In contrast, no binding proteins MORF2 and MORF9 through BiFC assays (Fig. 6B). between MBP-ECD1 and the rps14-80 oligonucleotide was The interaction with MORF proteins confirmed the effect of observed (Fig. 7B). ECD1 on RNA editing. To further test the ECD1–RNA interaction in vivo, RNAs co-immunoprecipitated with the anti-FLAG antibody were analysed by qRT-PCR using primers for transcripts containing ECD1 specifically interacts with the cis-elements of the editing sites of rps14-149, and including transcripts con- rps14-149 in vitro and in vivo taining rps14-80 and petL as controls. We detected enrichment There are two editing sites in rps14 transcripts, rps14-80 and fragments of rps14-149 in the anti-FLAG immupoprecipitate, rps14-149. In cotyledons of the ECD1-RNAi transgenic plants, but not of rps14-80 and petL (Fig.  8). This analysis suggested editing of rps14-149 was decreased dramatically, while that of that ECD1 also binds to a cis-element surrounding rps14-149 rps14-80 remained normal. If the mutation of ECD1 specif- in vivo. ically affects RNA-editing of rps14-149, then ECD1 should bind to a cis-element surrounding this editing site. To analyse ECD1 is indispensable for functional ribosomes in the ability of ECD1 to interact with such a cis-element, elec- plastids trophoretic mobility shift assays (EMSAs) were performed. The recombinant ECD1 protein with an N-terminal MBP tag was rps14 encodes the ribosomal protein S14 and is essential for expressed (Supplementary Fig. S7) and two oligoribonucleo- survival of tobacco plants (Tiller and Bock, 2014). Given the tides of 33 residues surrounding the editing sites of the rps14- extremely low editing efficiency of rps14-149 and the decreased 80 and rps14-149 transcripts were synthesized for analysis by accumulation of photosynthesis proteins in the ECD1-deficient Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 RNA-editing factor ECD1 and chloroplast development | 3045 Fig. 6. ECD1 interacts with multiple organellar RNA-editing factor (MORF) proteins. (A) Yeast two-hybrid screens with ECD1 showing that it interacts with MORF proteins. ECD1 was used as bait (BD), MORFs were used as prey (AD). The left panel shows the growth test on permissive medium lacking Trp –1 and Leu; the right panel shows the same clones on selective medium lacking Trp, Leu, and His, and containing 40 μg ml X-α-Gal (5-bromo-4-chloro- 3-indolyl-a-D-galactopyranoside). Yeast cells transformed with pGBKT7-53 and pGADT7-T were used as positive controls and cells transformed with pGBKT7-lam and pGADT7-T were negative controls. (B) Bimolecular fluorescence complementation (BiFC) assays showing that YFP -ECD1 interacts C C with MORF2-YFP (or MORF9-YFP ) to produce YFP fluorescence in the chloroplasts. The scale bar is 10 μm. (This figure is available in colour at JXB online.) mutant, we examined the levels of the RPS14 protein. The effects on ribosome levels, we examined the protein amounts level in the ECD1-RNAi transgenic plants decreased to less for two other plastid-encoded ribosomal proteins, RPS3 and than one-quarter of that in the wild-type in the cotyledons, RPL2. by immunoblot analysis. Both proteins were decreased but there was no change in the true leaves (Fig. 9A). RPS14 significantly in the ECD1-RNAi-1 line (Fig.  9B). Thus, the is required for the accumulation of ribosomal 30S subunits. To defect in the accumulation of RPS14 protein in the ECD1- determine whether the reduced content of RPS14 had any RNAi transgenic plants may have compromised ribosome Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 3046 | Jiang et al. Fig. 7. Gel electrophoretic mobility shift assays (EMSAs) with the rps14 oligoribonucleotides. (A) The sequences used as oligonucleotide probes. Asterisks indicate the editing sites. (B) EMSA showing that MBP-ECD1 binds to the sequence around rps14-149, but not rps14-80. Increasing concentrations of protein (25, 50 and 100 nM) were incubated with 10 nM probes. The positions of the shift and free probes are indicated. (C) Unlabeled oligoduplexes with 10-fold, 50-fold, and 100-fold excess were used for competition to confirm the specific interaction between ECD1 and the cis- element around rps14-149. (This figure is available in colour at JXB online.) Fig. 8. RNA immunoprecipitation analysis of ECD1 and rps14-149. (A) ECD1-FLAG protein accumulation in the 35S::ECD1-FLAG transgenic seedlings compared with the wild-type (WT). (B) RNA immunoprecipitation analysis. IP+, anti-FLAG immunoprecipitation; IP-, mock immunoprecipitation. rps14-80, rps14-149, and petL are fragments that contain the editing sites of rps14-80, rps14-149, and petL, respectively. Data are means (±SE) obtained from three replicates. accumulation, and the deficiency in translation was presumably unchanged, while Class  II genes (transcribed by both NEP responsible for the albino phenotype of cotyledons. and PEP) were differentially regulated in ECD1-RNAi-1. The transcript abundance of other chloroplast genes that are not clearly classified are shown in Supplementary Fig. S8. In order The ECD1 mutation affects plastid gene expression to verify these results, we carried out RNA blot analysis of the and plastid rRNA processing psbA, rbcL, clpP, and rpoA genes using sequence-specific labeled The expression of chloroplast genes significantly impacts on probes. The steady-state levels of transcripts were in almost chloroplast development. We examined the transcript abun- complete agreement with the qRT-PCR analysis, and the pro- dance of various chloroplast genes in 7-d-old seedlings by cessing patterns between the wild-type and ECD1-RNAi-1 qRT-PCR. The results showed that the transcript levels of plants did not differ greatly (Fig. 10B , Supplementary Fig. S9). Class I genes (transcribed preferentially by PEP) were signifi- These results indicated that ECD1 is essential for PEP but not cantly reduced in ECD1-RNAi-1 compared with the wild- for NEP activity. type (Fig. 10A). In contrast, transcript levels of Class III genes Decreased levels of rRNAs were found in the cotyle- (transcribed preferentially by NEP) were either increased or dons of the ECD1-RNAi transgenic plants using ethidium Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 RNA-editing factor ECD1 and chloroplast development | 3047 Fig. 9. Immunoblot of chloroplast ribosomal subunits in the wild-type (WT) and ECD1-RNAi-1 line. (A) Immunoblots of PRS14 in 7- and 14-d-old seedlings. C and T refer to the proteins in cotyledons and true leaves, respectively, in the 14-d samples. (B) Immunoblot of RPS3 and RPL2 in samples of whole seedlings at 7 d old. bromide-stained agarose gel assays (Fig.  10B, Supplementary especially embryo-lethal mutants. Chloroplast development is Fig.  S9). We next examined the role of ECD1 in rRNA closely related to embryo development and, as a consequence, metabolism. Chloroplast ribosomal RNAs are co-transcribed eliminating biosynthetic functions within the chloroplast and as a single RNA precursor that contains 16S, 23S, 4.5S, and 5S interfering with expression of the chloroplast genome often rRNAs, as well as two tRNAs (Fig. 11A). The precursor tran- results in embryo lethality in Arabidopsis (Bryant et al., 2011). script undergoes a complex series of processing events before A  set of 119 nuclear genes encoding chloroplast-localized maturation. The 23S–4.5S bi-cistronic RNA (3.2 kb) under- proteins has been identified, including many PPR proteins goes endonucleolytic cleavage to produce a mature 4.5S rRNA (Bryant et  al., 2011). Disruption of these genes results in an and a 23S precursor (2.9 kb), which undergoes further matu- embryo-defective phenotype, highlighting the importance ration and ultimately generates three species of 1.1, 1.3, and of chloroplasts in embryogenesis. Here, we identified a novel 0.5 kb. The 16S precursor RNA (1.7 kb) is processed to a 1.5- PPR protein, ECD1, which is required for chloroplast devel- kb mature 16S rRNA. RNA gel blot analysis was performed opment in seedlings. A  complete deletion of ECD1 resulted on the total leaf RNAs from 7- and 14-d-old wild-type and in embryo lethality, indicating that ECD1 is indispensable for ECD1-RNAi-1 plants using rRNA-specific probes (indicated plant growth and development. Evidence suggests that accD, in Fig.  11). Higher transcript levels of the 3.2-kb 23S–4.5S which encodes one subunit of a multimeric acetyl-CoA carb- rRNA precursor and the 1.7-kb 16S rRNA precursor were oxylase required for fatty acid biosynthesis, is amongst the most detected in the cotyledons of ECD1-RNAi-1, whereas the lev- important chloroplast genes required for embryo development els of the 1.5-kb 16S, 0.5-kb 23S, 0.1-kb 4.5S, and 0.12-kb in Arabidopsis (Bryant et al., 2011; Parker et al., 2016). Editing 5S mature rRNAs decreased drastically (Fig.  11B). However, of accD-794 and accD-58642 both decreased in the ECD1- no obvious differences in the levels of the rRNA transcripts RNAi transgenic plants (Supplementary Fig. S4). However, the between the wild-type and ECD1-RNAi-1 were observed in editing deficiency of these sites may have been caused indir- true leaves (Supplementary Fig. S10). On the other hand, the ectly by the albino phenotype since they were also affected in transcript levels of the two tRNAs encoded by this operon, trnI seedlings treated with lincomycin or spectinomycin. and trnA, accumulated to the same extent in ECD1-RNAi-1 In the ECD1-RNAi transgenic plants, the chloroplast ultra- and the wild-type cotyledons (Supplementary Fig. S11). These structure in cotyledons exhibited abnormal morphology and results suggested that ECD1 may play an important role in the thylakoid membrane structure was perturbed, suggesting plastid rRNA maturation. that the albino phenotype of cotyledons in these plants was probably due to developmentally defective chloroplasts. Proper accumulation of plastid ribosomal proteins is a prerequisite for assembling functional ribosomes and is necessary for chloro- Discussion plast development. A deficiency of nuclear-encoded chloroplast Chloroplasts are organelles that perform photosynthesis. The factors required for the synthesis of plastid ribosomal proteins development of a functional chloroplast is regulated by a large can compromise assembly and accumulation of chloroplast number of genetic factors, especially nuclear-encoded factors. ribosomes. PPR4 is required for the trans-splicing of the plas- Given their importance, extensive studies have been carried tid rps12 transcript and consequently affects the accumulation out focusing on the identification of nuclear genes essential of plastid ribosomes (Schmitz-Linneweber et  al., 2006). The for chloroplast development. However, numerous nuclear maize chloroplast protein PPR103 stabilizes the 5´-end of pro- mutants that impact on chloroplast development can easily cessed rpl16 mRNAs and a loss of plastid ribosomes was also be overlooked because of their severe, often lethal phenotype, detected in ppr103 mutants (Hammani et al., 2016). Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 3048 | Jiang et al. Fig. 10. Chloroplast gene expression in ECD1-RNAi-1 relative to the wild-type (WT). (A) Transcript levels of chloroplast genes (Classes I–III) were measured by quantitative RT-PCR (qRT-PCR). Data are given as log of ECD1-RNAi-1/WT ratios from at least three independent experiments. Class I genes refer to psbC, psbT, psbN, psbH, psbD, psbB, petD, psbA, ndhA, psbE, psbF, and rbcL. Class II genes refer to rrn23, rrn16, atpI, ycf1, clpP, rps16, ndhB, ndhF, atpE, psaJ, and rps18. Class III genes refer to accD, ycf2, rpoB, rpoC2, and rpoC1. (B) RNA blot analysis of transcript levels for the different chloroplast gene classes. An ethidium bromide-staining gel is shown as a loading control (EtBR). In cotyledons of the ECD1-RNAi transgenic plants, the complexes with other proteins (Hammani et al., 2009). RNA- accumulation of RPS14 decreased dramatically compared editing of rps14-149 changes Pro to Leu. Since Pro tends to with the wild-type, which may have been due to the edit- disrupt α-helices and thus leads to instability of proteins, edit- ing defects in rps14-149. We found that ECD1 is able to bind ing of this site could restore an α-helix and stabilize the RPS14 to the cis-element of rps14-149 but not to the other editing protein (Sugita et al., 2006). RPS14 is essential for the assem- site rps14-80, indicating that ECD1 specifically affects RNA- bly of the ribosomal 30S subunit and contributes to the pep- editing of rps14-149. RNA-editing defects result in amino tide environment of the peptidyl transferase center in E.  coli acid changes that may directly alter protein function, or act (Brochier et al., 2000), and is also essential in tobacco plastids by destabilizing the protein or by affecting its ability to form (Ahlert et  al., 2003; Tiller and Bock, 2014). In Physcomitrella Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 RNA-editing factor ECD1 and chloroplast development | 3049 Fig. 11. Expression and processing of chloroplast rRNA. (A) Schematic representation of the chloroplast rrn operon of Arabidopsis. The locations of probes (a–j) used for the RNA gel blot analysis and the size of transcripts (in kb) are indicated below the operon. (B) RNA gel blot analysis. Mature transcripts were decreased and the precursors were increased in ECD1-RNAi-1 compared to the wild-type (WT). RNA was extracted from 7-d-old seedlings. An ethidium bromide-staining gel is shown as a loading control (EtBr). patens, reduction of RNA editing in rps14-C2 impairs the specific defect in 16S rRNA maturation in a cotyledon-spe- translation of the RPS14 protein and affects the function of cific manner (Yamamoto et al., 2000). sco1 mutants are defec- the chloroplast ribosome, which then results in a pale-green tive in the chloroplast elongation factor G, which not only phenotype and decreased photosynthetic activity in PpPPR- affects chloroplast mRNA translation during chloroplast for- 45-RNAi plants (Ichinose et  al., 2014). Thus, the decrease of mation in cotyledons, but also other developmental processes RPS14 may block the proper assembly of plastid ribosomes such as germination and flowering (Albrecht et al., 2006). The in Arabidopsis. A deficiency in plastid ribosomes may account proteins affected by sco2/cyo1 are required for protein folding for the global defects in PEP-dependent transcripts, as transla- and both of them have DnaJ-like zinc finger domains (Shimada tion of core subunits of the plastid-encoded RNA polymer- et al., 2007; Albrecht et al., 2008). The expression of ECD1 is ase decreases to a level that is not sufficient for transcribing not limited to cotyledons only; however, the higher amount in the PEP-dependent genes (Hess et al., 1993; Zubko and Day, cotyledons and its specific role in RNA-editing of rps14-149 2002; Chi et al., 2008). As ribosome assembly and pre-rRNA in cotyledons but not in true leaves suggest that ECD1 is more processing are intimately linked, defects in rRNA processing important in cotyledons than in leaves and other organs. Since in the cotyledons of ECD1-RNAi-1 may also be the con- ECD1-RNAi-1 is not lethal, the editing of rps14-149 is not sequence of the deficiency in ribosome assembly (Charollais completely abolished. It is expected that more severe editing et  al., 2003; Granneman and Baserga, 2005). In addition, the defects occur in the ecd1 homozygotes, leading to an embryo- decreased accumulation of RPL2 and RPS3 provides fur- lethal phenotype. ther evidence for a deficiency of chloroplast ribosomes. Taken In conclusion, the results of our study indicate that the PPR together, insufficient accumulation of plastid ribosomes may protein ECD1 is a site-specific factor for the rps14-149 RNA- be the cause for the developmentally defective chloroplasts in editing site, and it is required for early chloroplast development cotyledons of ECD1-RNAi-1. in seedlings. A decrease in ECD1 expression leads to editing In the ECD1-RNAi transgenic plants, chloroplast develop- defects of rps14-149 in cotyledons, which result in decreased ment within cotyledons (but not in true leaves) was severely accumulation of the RPS14 protein; this in turn leads to impaired, leading to the formation of white cotyledons. In lower levels of plastid ribosomes in cotyledons, and thus to recent years, several mutants with a phenotype of cotyledon- defects in chloroplast development. Decreased expression of specific impairment have been isolated. wco mutants have a PEP-dependent genes and defective plastid rRNA processing Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 3050 | Jiang et al. Brochier C, Philippe H, Moreira D. 2000. The evolutionary history were also observed in ECD1-RNAi transgenic plants, which of ribosomal protein RpS14: horizontal gene transfer at the heart of the can largely be accounted for by limiting amounts of plastid ribosome. Trends in Genetics 16, 529–533. ribosomes. Bryant N, Lloyd J, Sweeney C, Myouga F, Meinke D. 2011. Identification of nuclear genes encoding chloroplast-localized proteins required for embryo development in Arabidopsis. Plant Physiology 155, 1678–1689. Cai W, Ji D, Peng L, Guo J, Ma J, Zou M, Lu C, Zhang L. 2009. LPA66 Supplementary data is required for editing psbF chloroplast transcripts in Arabidopsis. Plant Physiology 150, 1260–1271. Supplementary data are available at JXB online. Charollais J, Pflieger D, Vinh J, Dreyfus M, Iost I. 2003. The DEAD-box Fig. S1. Motifs and sequence analysis of ECD1. RNA helicase SrmB is involved in the assembly of 50S ribosomal subunits Fig. S2. Expression patterns of ECD1 within seedlings. in Escherichia coli. Molecular Microbiology 48, 1253–1265. Fig.  S3. Ultrastructure of chloroplasts in ECD1-RNAi-1 Chi W, He B, Manavski N, Mao J, Ji D, Lu C, Rochaix JD, Meurer and wild-type seedlings. J, Zhang L. 2014. RHON1 mediates a Rho-like activity for transcription termination in plastids of Arabidopsis thaliana. The Plant Cell 26, 4918–4932. Fig. S4. RNA-editing efficiency of var ious target sites in plas- Chi W, Ma J, Zhang D, Guo J, Chen F, Lu C, Zhang L. 2008. The tids of cotyledons in ECD1-RNAi-1 and wild-type seedlings, pentratricopeptide repeat protein DELAYED GREENING1 is involved in the and in seedlings treated with lincomycin or spectinomycin. regulation of early chloroplast development and chloroplast gene expression Fig. S5. RNA editing efficiency of rps14 transcripts in seed- in Arabidopsis. Plant Physiology 147, 573–584. lings treated with lincomycin or spectinomycin. Clough SJ, Bent AF. 1998. Floral dip: a simplified method for Fig.  S6. Plastid RNA editing in true leaves of ECD1- Agrobacterium-mediated transformation of Arabidopsis thaliana. The Plant Journal 16, 735–743. RNAi-1 and wild-type seedlings. Cushing DA, Forsthoefel NR, Gestaut DR, Vernon DM. 2005. Fig. S7. Purification of MBP-ECD1. Arabidopsis emb175 and other ppr knockout mutants reveal essential roles Fig. S8. Transcript levels of chloroplast genes for which the for pentatricopeptide repeat (PPR) proteins in plant embryogenesis. Planta transcribing RNA polymerase is unknown. 221, 424–436. Fig.  S9. RNA gel blot analysis of chloroplast RNAs in Granneman S, Baserga SJ. 2005. Crosstalk in gene expression: coupling and co-regulation of rDNA transcription, pre-ribosome assembly and pre- cotyledons and true leaves of 14-d-old wild-type and ECD1- rRNA processing. Current Opinion in Cell Biology 17, 281–286. RNAi-1 seedlings. Hammani K, Okuda K, Tanz SK, Chateigner-Boutin AL, Shikanai T, Fig.  S10. RNA gel blot analysis of rRNAs in cotyledons Small I. 2009. A study of new Arabidopsis chloroplast RNA editing mutants and true leaves of 14-d-old wild-type and ECD1-RNAi-1 reveals general features of editing factors and their target sites. The Plant Cell 21, 3686–3699. seedlings. Hammani K, Takenaka M, Miranda R, Barkan A. 2016. A PPR protein in Fig. S11. RNA gel blot analysis of trnI and trnA in cotyledons the PLS subfamily stabilizes the 5´-end of processed rpl16 mRNAs in maize and true leaves of wild-type and ECD1-RNAi-1 seedlings. chloroplasts. Nucleic Acids Research 44, 4278–4288. Table S1. List of primers used in this study. Hashimoto M, Endo T, Peltier G, Tasaka M, Shikanai T. 2003. A nucleus-encoded factor, CRR2, is essential for the expression of chloroplast ndhB in Arabidopsis. The Plant Journal 36, 541–549. Acknowledgements Hedtke B, Börner T, Weihe A. 1997. Mitochondrial and chloroplast phage-type RNA polymerases in Arabidopsis. Science 277, 809–811. We are grateful to Prof. Jean David Rochaix for revising the manu- Hess WR, Prombona A, Fieder B, Subramanian AR, Börner T. 1993. script. We thank Tiegang Lu (The Chinese Academy of Agricultural Chloroplast rps15 and the rpoB/C1/C2 gene cluster are strongly transcribed Sciences) for providing us with antibodies for the plastid-encoded ribo- in ribosome-deficient plastids: evidence for a functioning non-chloroplast- somal proteins RPS3 and RPL2. This study was supported by the State encoded RNA polymerase. The EMBO Journal 12, 563–571. Key Basic Research and Development Plan of China (2015CB150100) Ichinose M, Uchida M, Sugita M. 2014. Identification of a pentatricopeptide and the Youth Innovation Promotion Association CAS (2013059). repeat RNA editing factor in Physcomitrella patens chloroplasts. FEBS Letters 588, 4060–4064. Kim BH, Malec P, Waloszek A, von Arnim AG. 2012. Arabidopsis BPG2: References a phytochrome-regulated gene whose protein product binds to plastid ribosomal RNAs. Planta 236, 677–690. Ahlert D, Ruf S, Bock R. 2003. Plastid protein synthesis is required for plant development in tobacco. Proceedings of the National Academy of Kovtun Y, Chiu WL, Tena G, Sheen J. 2000. Functional analysis of Sciences, USA 100, 15730–15735. oxidative stress-activated mitogen-activated protein kinase cascade in plants. Proceedings of the National Academy of Sciences, USA 97, Albrecht V, Ingenfeld A, Apel K. 2006. Characterization of the snowy 2940–2945. cotyledon 1 mutant of Arabidopsis thaliana: the impact of chloroplast elongation factor G on chloroplast development and plant vitality. Plant Liere K, Maliga P. 2001. Plastid RNA polymerases in higher plants. In: Aro Molecular Biology 60, 507–518. E-M, Andersson B, eds. Regulation of photosynthesis. Dordrecht: Springer Netherlands, 29–49. Albrecht V, Ingenfeld A, Apel K. 2008. Snowy cotyledon 2: the identification of a zinc finger domain protein essential for chloroplast development in Lurin C, Andrés C, Aubourg S, et al. 2004. Genome-wide analysis of cotyledons but not in true leaves. Plant Molecular Biology 66, 599–608. Arabidopsis pentatricopeptide repeat proteins reveals their essential role in organelle biogenesis. The Plant Cell 16, 2089–2103. Aryamanesh N, Ruwe H, Sanglard LV, Eshraghi L, Bussell JD, Howell KA, Small I, des Francs-Small CC. 2017. The pentatricopeptide repeat Martínez-García JF, Monte E, Quail PH. 1999. A simple, rapid and protein EMB2654 is essential for trans-splicing of a chloroplast small quantitative method for preparing Arabidopsis protein extracts for ribosomal subunit transcript. Plant Physiology 173, 1164–1176. immunoblot analysis. The Plant Journal 20, 251–257. Barkan A, Small I. 2014. Pentatricopeptide repeat proteins in plants. Meinke DW. 1994. Seed development in Arabidopsis thaliana. In: Annual Review of Plant Biology 65, 415–442. Meyerowitz EM, Somerville CR, eds. Arabidopsis. New York: Cold Spring Harbor Laboratory Press, 253–259. Börner T, Aleynikova AY, Zubo YO, Kusnetsov VV. 2015. Chloroplast RNA polymerases: role in chloroplast biogenesis. Biochimica et Biophysica Mullet JE. 1993. Dynamic regulation of chloroplast transcription. Plant Acta 1847, 761–769. Physiology 103, 309–313. Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 RNA-editing factor ECD1 and chloroplast development | 3051 Parker N, Wang Y, Meinke D. 2016. Analysis of Arabidopsis accessions Sugita M, Miyata Y, Maruyama K, Sugiura C, Arikawa T, Higuchi M. hypersensitive to a loss of chloroplast translation. Plant Physiology 172, 2006. Extensive RNA editing in transcripts from the PsbB operon and RpoA 1862–1875. gene of plastids from the enigmatic moss Takakia lepidozioides. Bioscience, Biotechnology, and Biochemistry 70, 2268–2274. Pfalz J, Bayraktar OA, Prikryl J, Barkan A. 2009. Site-specific binding of a PPR protein defines and stabilizes 5´ and 3´ mRNA termini in chloroplasts. Sun X, Feng P, Xu X, Guo H, Ma J, Chi W, Lin R, Lu C, Zhang L. 2011. A The EMBO Journal 28, 2042–2052. chloroplast envelope-bound PHD transcription factor mediates chloroplast signals to the nucleus. Nature Communications 2, 477. Pryor KD, Leiting B. 1997. High-level expression of soluble protein in Escherichia coli using a His6-tag and maltose-binding-protein double- Tiller N, Bock R. 2014. The translational apparatus of plastids and its role affinity fusion system. Protein Expression and Purification 10, 309–319. in plant development. Molecular Plant 7, 1105–1120. Ruwe H, Castandet B, Schmitz-Linneweber C, Stern DB. 2013. Tseng CC, Lee CJ, Chung YT, Sung TY, Hsieh MH. 2013. Differential Arabidopsis chloroplast quantitative editotype. FEBS Letters 587, regulation of Arabidopsis plastid gene expression and RNA editing in non- 1429–1433. photosynthetic tissues. Plant Molecular Biology 82, 375–392. Sakamoto W, Miyagishima SY, Jarvis P. 2008. Chloroplast biogenesis: Tzafrir I, Pena-Muralla R, Dickerman A, et  al. 2004. Identification of control of plastid development, protein import, division and inheritance. The genes required for embryo development in Arabidopsis. Plant Physiology Arabidopsis Book 6, e0110. 135, 1206–1220. Sato S, Nakamura Y, Kaneko T, Asamizu E, Tabata S. 1999. Complete Walter M, Chaban C, Schütze K, et  al. 2004. Visualization of structure of the chloroplast genome of Arabidopsis thaliana. DNA Research protein interactions in living plant cells using bimolecular fluorescence 6, 283–290. complementation. The Plant Journal 40, 428–438. Schmitz-Linneweber C, Small I. 2008. Pentatricopeptide repeat proteins: a Wang Y, Ren Y, Zhou K, et al. 2017. WHITE STRIPE LEAF4 encodes a socket set for organelle gene expression. Trends in Plant Science 13, 663–670. novel P-type PPR protein required for chloroplast biogenesis during early leaf development. Frontiers in Plant Science 8, 1116. Schmitz-Linneweber C, Williams-Carrier RE, Williams-Voelker PM, Kroeger TS, Vichas A, Barkan A. 2006. A pentatricopeptide repeat Yamamoto YY, Puente P, Deng XW. 2000. An Arabidopsis cotyledon- protein facilitates the trans-splicing of the maize chloroplast rps12 pre- specific albino locus: a possible role in 16S rRNA maturation. Plant & Cell mRNA. The Plant Cell 18, 2650–2663. Physiology 41, 68–76. Shikanai T, Fujii S. 2013. Function of PPR proteins in plastid gene Zhou W, Lu Q, Li Q, Wang L, Ding S, Zhang A, Wen X, Zhang L, Lu C. expression. RNA Biology 10, 1446–1456. 2017. PPR-SMR protein SOT1 has RNA endonuclease activity. Proceedings Shimada H, Mochizuki M, Ogura K, et al. 2007. Arabidopsis cotyledon- of the National Academy of Sciences, USA 114, E1554–E1563. specific chloroplast biogenesis factor CYO1 is a protein disulfide isomerase. Zubko MK, Day A. 2002. Differential regulation of genes transcribed by The Plant Cell 19, 3157–3169. nucleus-encoded plastid RNA polymerase, and DNA amplification, within Small ID, Peeters N. 2000. The PPR motif – a TPR-related motif prevalent ribosome-deficient plastids in stable phenocopies of cereal albino mutants. in plant organellar proteins. Trends in Biochemical Sciences 25, 46–47. Molecular Genetics and Genomics 267, 27–37. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Experimental Botany Oxford University Press

ECD1 functions as an RNA-editing trans-factor of rps14-149 in plastids and is required for early chloroplast development in seedlings

Loading next page...
1
 
/lp/ou_press/ecd1-functions-as-an-rna-editing-trans-factor-of-rps14-149-in-plastids-xLG7mEU07z

References (54)

Publisher
Oxford University Press
Copyright
Copyright © 2022 Society for Experimental Biology
ISSN
0022-0957
eISSN
1460-2431
DOI
10.1093/jxb/ery139
Publisher site
See Article on Publisher Site

Abstract

Chloroplast development is a highly complex process and the regulatory mechanisms have not yet been fully charac- terized. In this study, we identified Early Chloroplast Development 1 (ECD1), a chloroplast-localized pentatricopeptide repeat protein (PPR) belonging to the PLS subfamily. Inactivation of ECD1 in Arabidopsis led to embryo lethality, and abnormal embryogenesis occurred in ecd1/+ heterozygous plants. A decrease in ECD1 expression induced by RNAi resulted in seedlings with albino cotyledons but normal true leaves. The aberrant morphology and under-developed thylakoid membrane system in cotyledons of RNAi seedlings suggests a role of ECD1 specifically in chloroplast development in seedlings. In cotyledons of ECD1-RNAi plants, RNA-editing of rps14-149 (encoding ribosomal protein S14) was seriously impaired. In addition, dramatically decreased plastid-encoded RNA polymerase-dependent gene expression and abnormal chloroplast rRNA processing were also observed. Taken together, our results indicate that ECD1 is indispensable for chloroplast development at the seedling stage in Arabidopsis. Keywords: Arabidopsis, chloroplast development, cotyledon, Early Chloroplast Development 1 (ECD1), early stages, pentatricopeptide repeat protein (PPR), RNA editing. Introduction Chloroplasts are not only the exclusive organelles that perform polymerases: plastid-encoded bacterial-type RNA polymer- photosynthesis but they are also responsible for many other ase (PEP) and nuclear-encoded phage-type RNA polymerase biosynthetic processes, such as the synthesis of amino acids, (NEP) (Hedtke et  al., 1997; Liere and Maliga, 2001; Börner hormones, and metabolites (Sakamoto et al., 2008). The devel- et al., 2015). Given their increasing activity during chloroplast opment of functional chloroplasts is a prerequisite for photo- development, PEPs are clearly crucial for chloroplast develop- synthesis and is also tightly co-ordinated with plant growth and ment at early stages of plant growth (Mullet, 1993). During development. Chloroplast gene expression is crucial for chlo- chloroplast biogenesis, plastid ribosomal proteins are required roplast development, and is carried out by two kinds of RNA to establish a functional chloroplast translational apparatus and © The Author(s) 2018. Published by Oxford University Press on behalf of the Society for Experimental Biology. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 3038 | Jiang et al. deficiency of these proteins leads to lethality (Tiller and Bock, and which belongs to the PLS subgroup of the PPR family. 2014). The successful assembly of ribosomal proteins in chloro- Disruption of ECD1 leads to embryo lethality and RNAi lines plast is therefore very important for chloroplast development. display albino cotyledons but normal leaves. Aberrant chloro- Derived through endosymbiosis from cyanobacteria, chlo- plast ultrastructure and deficient RNA-editing of rps14-149 roplasts are semi-autonomous organelles that have their own (encoding ribosomal protein S14) in plastids were detected in genome. However, there are only about 100–150 genes in the cotyledons of the ECD1-RNAi transgenic plants. Mutation of plastid genome, the products of which are mainly involved in ECD1 also resulted in decreased expression of PEP-dependent photosynthesis and plastid gene expression (Sato et al., 1999). genes and abnormal rRNA processing. Our results indicate The vast majority of chloroplast proteins (>2000) are encoded that ECD1 plays a vital role in chloroplast development in in the nucleus, translated in the cytosol, and then imported seedlings. into the chloroplast. Thus, the formation of functional chlo- roplasts relies on co-ordination of gene expression between Materials and methods the plastid and the nucleus. Chloroplast gene expression is regulated by a set of nuclear-encoded factors. Among these, Plant materials and growth conditions pentatricopeptide repeat (PPR) proteins, which constitute one We obtained the T-DNA insertion line CS16045 (ecotype Columbia) of the largest protein family in land plants, have been dem- from the Arabidopsis Biological Resource Center (https://abrc.osu.edu/; onstrated to play important roles in chloroplast gene expres- last accessed 27 Apr il 2018). This line is the same as the one used in previous sion and function. Mutations of PPR genes usually result in studies by Tzafrir et al. (2004) and Cushing et al. (2005). The T -DNA inser- tion was confirmed by PCR with T-DNA-specific primers. Seeds of wild- seedling-lethal or embryo-lethal phenotypes. The Arabidopsis type and mutant plants were surface-sterilized after incubation at 4 °C for genome encodes more than 450 members of this family (Lurin 3 d to synchronize germination, then sown on Murashige and Skoog (MS) et  al., 2004; Shikanai and Fujii, 2013), and almost all of them medium containing 2% (w/v) sucrose. Plants were grown in soil under a –2 –1 are predicated to localize to plastids or mitochondria (Lurin 12/12-h light/dark cycle with a photon flux density of 120 μmol m s et  al., 2004). Members of the family are characterized by the at 22  °C. For lincomycin and spectinomycin treatment, the wild-type seeds were surface-sterilized and placed on MS media with the addition PPR motif, which appear as tandem repeats of a highly degen- –1 of 500 μM lincomycin or 50 mg l spectinomycin. The seedlings were erate unit of 35 amino acids (Small and Peeters, 2000; Lurin harvested after 7 d. et  al., 2004). The PPR protein family is classified into P and To produce ECD1 knock-down plants, an RNAi construct for ECD1 PLS subfamilies (Lurin et al., 2004), the latter being specific to was generated. A fragment of 434 bp of the ECD1 gene (from nucleotides land plants. The P subfamily usually does not contain any other 286 to 719) was amplified and inserted into the PFGC5941 vector. The forward restriction endonucleases were NcoI and SwaI, and the reverse conserved motifs except for the canonical PPR (P) motifs. By enzymes were XbaI and BamHI. The constructs were transformed into contrast, the PLS subfamily contains long (L) and short (S) Agrobacterium tumefaciens strain GV3101 and introduced into the wild- PPR-like motifs as well as classic PPR motifs. In addition, type plants by the floral dip method (Clough and Bent, 1998). Transgenic –1 based on the presence of different C-terminal motifs, the PLS plants were selected on MS medium containing 50 μg ml Basta. subfamily is further divided into the PLS, E, E+, and DYW subgroups (Schmitz-Linneweber and Small, 2008). Subcellular localization of GFP proteins PPR proteins have been reported to be involved in almost DNA encoding the 218 N-terminal amino acids of ECD1 was ampli- all stages of chloroplast gene expression. For example, PPR10 fied and ligated into the green fluorescent protein (GFP) fusion vector is required for the accumulation of processed RNAs with the pUC18-35S-sGFP with GFP as a reporter. The controls with mitochon- 5´ or 3´ terminus in the atpI-atpH or psaf-rpl33 intercistronic drial-, chloroplast-, and nuclear-localization signals were FROSTBITE1 region (Pfalz et al., 2009; Barkan and Small, 2014). CRR2 was (FRO1), ribulose bisphosphate carboxylase small subunit (RbcS), and the first reported DYW-PPR protein and it is involved in the PTM-N (Sun et  al., 2011), respectively. The resulting fusion constructs and the control vectors were introduced into Arabidopsis mesophyll pro- intergenic RNA cleavage between rps7 and ndhB (Hashimoto toplasts according to the PEG-mediated method (Kovtun et  al., 2000). et  al., 2003; Shikanai and Fujii, 2013). There are ppr mutants Fluorescence analysis was performed on an LSM 510 Meta confocal laser that affect PEP-dependent gene expression, such as dg1 (Chi scanning system (LSM510; Carl Zeiss, Jena, Germany). et al., 2008). SOT1, a PPR protein with a small MutS-related (SMR) domain has endonuclease activity. Its PPR domain spe- cifically recognizes a 13-nucleotide RNA sequence in the 5´ RNA gel blotting, RT-PCR, and quantitative RT-PCR end of the chloroplast 23S-4.5S rRNA precursor (Zhou et al., Total leaf RNA was extracted from 7-d-old seedlings, and from 14-d-old cotyledons and true leaves using an RNeasy Plant Mini kit (Qiagen). 2017). A r ice mutant, wsl, in which a PPR protein WSL is miss- RNA concentration was determined using thermo NanoDrop 2000. ing, exhibits reduced translation efficiency caused by abnormal Total RNA from seedlings of the wild-type and the ECD1-RNAi-1 splicing of the rpl2 gene (Wang et al., 2017). The PPR protein line was separated on 1.3% (w/v) agarose-formaldehyde gels, blotted to a EMB2654 is required for the trans-splicing of the plastid rps12 nylon membrane, and subsequently hybridized with a probe labeled with transcript and its binding site is localized on one of the intron P. The probes were prepared by PCR amplification and labeled using the Prime-a-Gene Labeling System (SGMB01-Promega-U1100). halves (Aryamanesh et al., 2017). RNA was used to generate first-strand cDNA in a 20-μl reaction using Despite the presence of numerous PPR proteins in higher the Superscript III cDNA synthesis system (Invitrogen). The resulting plants, their functions in the regulation of chloroplast devel- cDNA samples were used as templates for RT-PCR analysis. Quantitative opment has not yet been elucidated. Here, we report a novel RT-PCR was performed using the SYBR Premix ExTaq Kit (Takara) chloroplast factor, Early Chloroplast Development 1 (ECD1), following the manufacturer’s instructions with a Light Cycler 480 system. The expression level was normalized to that of an ACTIN control. that is involved in early chloroplast development in seedlings Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 RNA-editing factor ECD1 and chloroplast development | 3039 Histochemical GUS staining Yeast two-hybrid assays Tissues were incubated in cold 90% (v/v) acetone and placed in a staining The CDS of ECD1 without the first 150 bp encoding the transit pep- buffer (100 mM sodium phosphate buffer, pH 7.2, 0.2% Triton X-100, tide was amplified by PCR and cloned into pGBKT7 DNA-BD as bait. 10 mM potassium ferrocyanide, 10 mM potassium ferricyanide, 0.25M Sequences encoding the mature MORF2, MORF3, MORF6, MORF8, EDTA, and 1  mM X-gluc). After vacuum-infiltration for 15–30  min, and MORF9 proteins were cloned into pGADT7 as prey. The prey samples were incubated for 16–24 h at 37 °C. β-Glucuronidase (GUS)- and bait constructs were co-transformed into Y2HGold yeast cells. The stained tissues were cleared with an ethanol series of 20% (v/v), 30% transformation was performed using the Matchmaker Gold Yeast Two- (v/v), and 50% (v/v) for 30 min in turn, and finally incubated in a solu- Hybrid System (Clontech) according to the manufacturer’s instructions. tion of 70% (v/v) ethanol and 30% (v/v) acetic acid for at least 30 min Interactions were determined by growing diploid yeast colonies on syn- –1 until the tissues became transparent enough to observe under a dissecting thetic dropout (SD) medium containing 40 μg ml X-α-Gal (5-bromo- Olympus SZX16 microscope. 4-chloro-3-indolyl-α-d-galactopyranoside) without tryptophan, leucine, and histidine. Protein isolation and immunoblot analysis RNA immunoprecipitation assays Total proteins were prepared as previously described (Martínez-García et al., 1999). Protein concentrations were determined using the Bio-Rad RIP assays were performed as described previously (Kim et  al., 2012) DC protein assay. For immunoblot analysis, total proteins were separated using 7-d-old 35S::ECD1-FLAG transgenic seedlings. Anti-FLAG by SDS-PAGE and transferred to nitrocellulose membranes. The mem- M2 magnetic beads were obtained from Sigma (M8823). Protein A/G branes were incubated with specific primary antibodies, and the signals Sepharose incubated with pre-immune serum was used as the control. were detected using a Pro-Light HRP Chemiluminescent Kit (Tiangen RNA was isolated by phenol-chloroform isoamyl alcohol extraction and Biotech). PsaA, D1, LHCII, Cytb6, Cytf, CF0II, and RPS14 were then analysed by qRT-PCR. expressed and purified in-house at our laboratory and used to generate polyclonal antibodies in rabbits. The antisera for RPL2 and RPS3 were Electrophoretic mobility shift assays provided by Tiegang Lu, and the anti-FLAG antibody was obtained from Abmart (www.ab-mart.com/; last accessed 27 April 2018). EMSAs were carried out using a LightShift Chemiluminescent RNA EMSA Kit (Thermo 20158). After incubation for 20 min at 25 °C, the samples were resolved on a 6% Tris-borate gel in 0.5× TBE buffer, trans- Analysis of RNA editing ferred to a nylon membrane, and subsequently processed using a chemi- A series of specific primers were used to amplify the regions of the genes luminescent detection kit (Thermo 89880). The 5´-end biotin-labeled containing the editing sites in Arabidopsis (Cai et  al., 2009) from the oligoribonucleotides rps14-80 and rps14-149 were synthesized and cDNA using RT-PCR, and the products were sequenced directly (for a list labeled by Takara Bio Inc. To produce recombinant MBP-ECD1 pro- of primers used in this study see Supplementary Table S1 at JXB online). teins, the coding sequence of ECD1 lacking the transit peptide sequence The levels of RNA editing were estimated by the relative heights of the was PCR-amplified, digested with SacI and NotI, and inserted into pET - peaks of the nucleotide in the sequence analysed. Plasmids prepared from MALc-H (Pryor and Leiting, 1997). Recombinant protein was expressed approximately 90 independent colonies of each sample were sequenced to and then purified by amylose affinity chromatography according to the determine the RNA-editing efficiency of rps14-80 and rps14-149. manufacturer’s instructions (New England BioLabs). Transmission electron microscopy Results For TEM processing, wild-type and ECD1-RNAi-1 leaves from 7-d-old plants, and from cotyledons and true leaves from 14-d-old plants were Mutations in ECD1 produce defects in embryogenesis collected. The tissue was cut into small pieces and fixed in 3% glutaral- dehyde in phosphate buffer for 4 h at 4 °C. After fixation, the tissue was To study the detailed mechanisms of chloroplast develop- rinsed in phosphate buffer 3–4 times and then post-fixed in 1% OsO ment, we obtained a series of T-DNA insertion lines from the overnight at 4  °C. After rinsing in phosphate buffer again, the samples Arabidopsis Biological Resource Center, the products of which were dehydrated in an ethanol series, infiltrated with a graded series of are predicted to be candidates for chloroplast biogenesis fac- epoxy resin in epoxy propane, and embedded in Epon 812 resin. Thin sections were obtained using a diamond knife on a Reichert OM2 ultra- tors. An embryo-lethal line CS16045 of the gene AT3G49170, microtome, stained with 2% uranylacetate, pH5.0, followed by 10  mM designated as ecd1 (originally called emb2261), attracted our lead citrate, pH12, and viewed with a transmission electron microscope attention for further investigation. Failure to identify any prog- (JEM-1230; JEOL). eny homozygous for the ecd1 mutant allele suggested that the mutation causes embryonic lethality. We dissected the develop- Analysis of embryo development ing siliques and assessed the seeds under a dissecting micro- Embryos were excised from wild-type and ecd1+/ siliques at different scope. In wild-type siliques, all the ovules developed normally, developmental stages and cleared in Hoyer’s solution (7.5 g gum arabic, while in the heterozygous ecd1/+ siliques, some ovules were 100 g chloral hydrate, and 5 ml glycerol in 30 ml water) as described by white (Fig. 1A). In 33 siliques from the heterozygous ecd1/+ Meinke (1994). Embryo development was studied microscopically using plants, 177 out of 726 ovules were white, making the ratio an Olympus BH-2 microscope equipped with Nomarski optics. of white to green ovules 1:3 (χ =0.166, P>0.05) (data not shown). In older siliques, the white ovules became shrunken Bimolecular fluorescence complementation assays and aborted. BiFC assays were performed as previously described (Walter et  al., 2004). To deter mine precisely the stage of embryogenesis dur ing Full-length cDNA of ECD1 was cloned into pSAT4A-nEYFP-N1, and full- which the ECD1 mutant arrested development, develop- length cDNAs of MORF2 and MORF9 (multiple organellar RNA-editing ing seeds at various stages from self-pollinated heterozygous factor) were cloned into pSAT4A-cEYFP-N1. The plasmids were co-trans- plants were cleared and observed using differential interfer- formed into protoplasts. Yellow fluorescent protein (YFP) was imaged using a confocal laser scanning microscope (LSM510; Carl Zeiss, Jena, Germany). ence contrast microscopy. The normal wild-type embryos Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 3040 | Jiang et al. underwent typical developmental stages, ranging from pre- The ECD1 gene encodes a chloroplast PPR protein globular, globular, heart-shaped, torpedo-shaped, through belonging to the PLS subfamily to cotyledon and maturity. However, in the heterozygous The ECD1 gene encodes a putative protein of 850 amino acids ecd1/+ siliques, although the homozygous mutant embryos with a predicated molecular mass of 95.5 kDa. The N-terminal consistently initiated cotyledons and showed continued 50 amino acids are predicated by ChloroP1.1 (http://www.cbs. growth and cell division, beyond the heart stage the devel- dtu.dk/services/ChloroP/; last accessed 27 April 2018) to con- opment of the embryos was significantly slower than that stitute a chloroplast transit peptide. Sequence analysis revealed of the wild-type (Fig.  1B). Mutant embryos consistently that the ECD1 protein contains 17 PPR or PPR-like (P, L, failed to elongate, developing instead as v-shaped embryos and S) motifs, together with one E motif, one E+ motif, and with wide, stunted cotyledons and no hypocotyl (see also one DYW motif in the C-terminal part (see Supplementary Cushing et al., 2005). Fig. S1A). It belongs to the PLS subgroup of the PPR protein family. Protein alignment showed that ECD1 shares significant identity at the amino acid level with proteins from Brassica, Knock-down of ECD1 results in a cotyledon-specific grape (Vitis), eggplant (Solanum), Zea mays, and rice (Oryza) albino phenotype (Supplementary Fig. S1B). To determine the subcellular localization of the ECD1 pro- To further investigate the function of ECD1, we constructed tein, the 218 N-terminal amino acids were fused to the N RNAi lines. A  total of 42 out of 76 RNAi-ECD1 trans- terminus of synthetic GFP (sGFP). The ECD1-GFP fusion genic lines with the abnormal cotyledon phenotype were protein was transiently expressed in Arabidopsis protoplasts obtained (data not shown). In further studies, three RNAi under the control of the cauliflower mosaic virus 35S pro- transgenic lines with a range of stable phenotypes with moter. We observed that the GFP fluorescence merged with respect to white cotyledons and stunted plant growth were the chlorophyll autofluorescence (Fig.  3A), indicating that selected (Fig. 2A, B); however, these lines all had true leaves ECD1 is a chloroplast protein. When ECD1-GFP was tran- that were normal green. RT-PCR showed that the pheno- siently co-expressed with red fluorescent protein (RFP) fused types of these RNAi lines correlated with the expression with pTAC5 (a well-characterized protein known to localize levels of the ECD1 gene (Fig.  2C). These results indicated in nucleoids; Chi et al., 2014), the green and red fluorescence that disruption of the ECD1 gene led to abnormal cotyle- signals within the chloroplasts were found to merge, indicat- dons. The most severely affected line, ECD1-RNAi-1, in ing that ECD1 and pTAC5 were co-localized in chloroplast which the cotyledons were albino, was selected for further nucleoids (Fig.  3B). To further determine the localization of analysis. Fig. 1. Embryogenesis of wild type and ecd1/+ embryos. (A) A heterozygous ecd1/+ mutant silique showing that approximately one-quarter of the ovules are albino compared with the wild-type (WT). Scare bars are 0.5 mm. (B) I–VI, normal embryos of the wild-type: I, pre-globular; II, globular; III, heart-shaped; IV, torpedo-shaped; V, cotyledon; VI, mature. VII–XII, embryos of ecd1/+. VII–IX, embryos are similar to the wild-type; X–XII, development is arrested, with v-shaped embryos with wide, stunted cotyledons and no hypocotyl. Scare bars are 20 μm. (This figure is available in colour at JXB online.) Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 RNA-editing factor ECD1 and chloroplast development | 3041 Fig. 2. Characterization of the ECD1-RNAi transgenic plants. (A) Identification and isolation of RNAi lines with different degrees of inhibition of ECD1 expression (WT: wild-type, ecotype Columbia). Plants were grown on MS medium with 2% (w/v) sucrose for 7 d. Scare bars are 1 mm. (B) Albino cotyledon phenotype of ECD1-RNAi-1 compared with the WT. Plants were grown on MS medium with 2% (w/v) sucrose (3–14 d) or soil (28 d). Scare bars are 1 mm for 3–14 d; 1 cm for 28 d. (C) Reverse transcription PCR (RT-PCR) using specific primers for AT3G49170 or Actin12 for 27 cycles for WT and RNAi lines with different degrees of inhibition of ECD1. (This figure is available in colour at JXB online.) ECD1, intact chloroplasts of overexpressing ECD1-Flag trans- In addition, GUS activity was also detected in rosette leaves, genic plants were isolated and fractionated, and the proteins flower buds, flowers, and siliques, with minimal expression were separated by SDS-PAGE followed by immunoblot anal- also observed in roots (Supplementary Fig. S2). In the flowers, ysis using an anti-FLAG antibody. The ECD1 protein was GUS staining was observed exclusively in green tissues, such as sepals, stamens, and carpels, but not in petals. Taken together, detected in both the stromal and thylakoid fractions (Fig. 3C). these findings showed that ECD1 was widely expressed throughout the plant, but the highest expression was in the Gene expression pattern of ECD1 cotyledons, which was consistent with the albino cotyledon To investigate the expression pattern of the ECD1 gene in phenotype of the ECD1-RNAi transgenic plants. The results Arabidopsis, we made transgenic plants expressing the GUS also indicated that ECD1 expression is developmentally con- protein under the control of the ECD1. The highest expres- trolled and corresponds to early chloroplast development in sion levels were observed in the cotyledons of seedlings. seedlings. Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 3042 | Jiang et al. Fig. 3. ECD1 is localized in the chloroplasts. (A) Subcellular localization of ECD1-GFP. The fluorescence of the ECD1-GFP fusion protein in protoplasts was observed using confocal laser scanning microscopy. Green fluorescence signals, chlorophyll red autofluorescence signals, and merged images are shown. The controls used for transformations are indicated to the right: Mit-GFP, control with mitochondrial localization signal of FROSTBITE1 (FRO1); Chl-GFP, control with the transit peptide of the Rubisco small subunit (RbcS); Nuc-GFP, control with nuclear localization signal of PTM-N (Sun et al., 2011). The scare bar is 10 μm. (B) Co-localization of ECD1-GFP with the pTAC5-RFP protein in chloroplast nucleoids. The red fluorescence of chlorophyll has been adjusted for better contrast. The scare bar is 10 μm. (C) Immunoblot analysis of the ECD1-FLAG fusion protein in chloroplast subfractions. ECD1 localizes to both the stroma and thylakoid membrane fractions. Intact chloroplasts were isolated from 35S::ECD1-FLAG transgenic seedlings and then separated into fractions. Polyclonal antisera were used to detect the ECD1-FLAG fusion protein, the light-harvesting complex II (LHCII), and the Rubisco large subunit (rbcL). (This figure is available in colour at JXB online.) in seedlings. We next examined chloroplast morphology and The ECD1 mutation affects early chloroplast ultrastructure in the transgenic plants using TEM. The wild-type development in seedlings chloroplast ultrastructure was similar in cotyledons and true Together with the chloroplast localization of the ECD1 protein, leaves, with well-developed thylakoid membranes composed of the albino cotyledons of ECD1-RNAi-1 plants suggested that grana connected by stroma lamellae (Supplementary Fig.  S3). the function of ECD1 is related to early chloroplast development Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 RNA-editing factor ECD1 and chloroplast development | 3043 The chloroplasts of albino cotyledons of ECD1-RNAi-1 had of cotyledons and true leaves in the wild-type and transgenic no organized thylakoid membrane system but instead contained plants; however, we did not include the nine sites recently a large number of round or oblong membrane-bound internal found by Ruwe et al. (2013) because of their extremely low vesicles. However, the chloroplasts of true leaves from ECD1- editing efficiency (<10%) even in the wild-type. The results RNAi-1 contained organized thylakoid membranes similar showed that the editing efficiency of the ribosomal protein to those of the wild type. The results indicated that ECD1 is rps14-149 decreased to extreme low levels in the cotyledons involved in chloroplast development in seedlings. of ECD1-RNAi-1 compared to that of the wild-type (Fig. 5). Multiple other sites were also affected to varying degrees (Supplementary Fig.  S4), but no significant differences in The ECD1 mutation affects the accumulation of the editing efficiency were detected in true leaves between proteins of photosynthetic complexes ECD1-RNAi-1 and the wild-type (Supplementary Fig. S6). Since the chloroplast ultrastructure was affected in the ECD1- To test whether the editing deficiency was indirectly caused RNAi transgenic plants, we investigated whether the accumu- by the albino cotyledon phenotype, we pharmacologically lation of proteins of the photosynthetic complexes was different induced albinism by using the plastid translation inhibitor in these plants. Immunoblotting was performed to analyse the lincomycin. Lincomycin treatment results in a severe albino levels of the proteins of each of the thylakoid protein com- phenotype and it has been reported to have severe effects on plexes. The protein levels of PSI (PsaA), PSII (D1 and LHCII), RNA editing (Tseng et al., 2013). Our results showed that in Cytb6/f (Cytb6 and Cytf), and ATPase (CF0II) of cotyledons lincomycin-treated seedlings, the editing efficiency of many in ECD1-RNAi-1 were dramatically decreased compared to sites was significantly reduced or even completely abolished, those of the wild type (Fig. 4). However, the accumulation of including accD-794, accD-58642, petL-5, ndhB-836, ndhD-878, these photosynthetic proteins in true leaves of ECD1-RNAi-1 and ndhF-290 and these were also affected in ECD1-RNAi-1 was similar to that in the wild-type. (Supplementary Fig. S4). However, no obvious editing defi- ciency of rps14-149 was detected in lincomycin-treated seed- lings (Supplementary Figs S4, S5). To rule out the possibility ECD1 is required for RNA-editing of rps14-149 in that the editing defects were unique to lincomycin, we eval- plastids of cotyledons uated the effect of another inhibitor of chloroplast transla- Previous studies have shown that the critical function of the tion, spectinomycin, on editing and obtained similar results DYW domain is involved in RNA editing. To test whether (Figs  S4, S5). We therefore concluded that the mutation of the RNA-editing status in the ECD1-RNAi transgenic plants ECD1 specifically affected RNA editing of rps14-149 in plas- was altered, we sequenced the 34 known editing sites in plastids tids of cotyledons. Fig. 4. Immunoblot analysis of photosynthetic proteins. Total protein was separated by 10% Tricine/SDS-PAGE, electro-blotted, and probed using specific anti-PsaA, anti-D1, anti-LHCII, anti-Cytb6, anti-Cytf, and anti-CF0II antibodies. C and T refer to the proteins in cotyledons and true leaves, respectively, of ECD1-RNA-1 and wild-type (WT) seedlings. Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 3044 | Jiang et al. Fig. 5. Analysis of RNA-editing of rps14 transcripts from wild-type (WT) and ECD1-RNAi-1 seedlings. (A) RT-PCR products containing the rps14-80 and rps14-149 editing sites were directly sequenced. The editing sites of rps14-80 and rps14-149 are indicated by asterisks above the corresponding peaks. (B) The editing efficiency was determined by analysis of approximately 90 independent clones from RT-PCR products for rps14-80 and rps14-149 in the wild-type and ECD1-RNAi-1 line. (This figure is available in colour at JXB online.) Yeast two-hybrid screening with ECD1 as bait identified EMSA (Fig. 7A). As shown in Fig. 7B, binding of MBP-ECD1 several members of the family of multiple organellar RNA- to the rps14-149 oligonucleotide was increased with increasing editing factors (MORF) including MORF2, MORF3, concentrations of the MBP-ECD1 protein. The specificity of MORF6, MORF8, and MORF9 (Fig. 6A). We also confirmed binding was confirmed using the same unlabeled oligoribo- the interactions between ECD1 and the chloroplast-targeted nucleotide as a competitor (Fig.  7C), In contrast, no binding proteins MORF2 and MORF9 through BiFC assays (Fig. 6B). between MBP-ECD1 and the rps14-80 oligonucleotide was The interaction with MORF proteins confirmed the effect of observed (Fig. 7B). ECD1 on RNA editing. To further test the ECD1–RNA interaction in vivo, RNAs co-immunoprecipitated with the anti-FLAG antibody were analysed by qRT-PCR using primers for transcripts containing ECD1 specifically interacts with the cis-elements of the editing sites of rps14-149, and including transcripts con- rps14-149 in vitro and in vivo taining rps14-80 and petL as controls. We detected enrichment There are two editing sites in rps14 transcripts, rps14-80 and fragments of rps14-149 in the anti-FLAG immupoprecipitate, rps14-149. In cotyledons of the ECD1-RNAi transgenic plants, but not of rps14-80 and petL (Fig.  8). This analysis suggested editing of rps14-149 was decreased dramatically, while that of that ECD1 also binds to a cis-element surrounding rps14-149 rps14-80 remained normal. If the mutation of ECD1 specif- in vivo. ically affects RNA-editing of rps14-149, then ECD1 should bind to a cis-element surrounding this editing site. To analyse ECD1 is indispensable for functional ribosomes in the ability of ECD1 to interact with such a cis-element, elec- plastids trophoretic mobility shift assays (EMSAs) were performed. The recombinant ECD1 protein with an N-terminal MBP tag was rps14 encodes the ribosomal protein S14 and is essential for expressed (Supplementary Fig. S7) and two oligoribonucleo- survival of tobacco plants (Tiller and Bock, 2014). Given the tides of 33 residues surrounding the editing sites of the rps14- extremely low editing efficiency of rps14-149 and the decreased 80 and rps14-149 transcripts were synthesized for analysis by accumulation of photosynthesis proteins in the ECD1-deficient Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 RNA-editing factor ECD1 and chloroplast development | 3045 Fig. 6. ECD1 interacts with multiple organellar RNA-editing factor (MORF) proteins. (A) Yeast two-hybrid screens with ECD1 showing that it interacts with MORF proteins. ECD1 was used as bait (BD), MORFs were used as prey (AD). The left panel shows the growth test on permissive medium lacking Trp –1 and Leu; the right panel shows the same clones on selective medium lacking Trp, Leu, and His, and containing 40 μg ml X-α-Gal (5-bromo-4-chloro- 3-indolyl-a-D-galactopyranoside). Yeast cells transformed with pGBKT7-53 and pGADT7-T were used as positive controls and cells transformed with pGBKT7-lam and pGADT7-T were negative controls. (B) Bimolecular fluorescence complementation (BiFC) assays showing that YFP -ECD1 interacts C C with MORF2-YFP (or MORF9-YFP ) to produce YFP fluorescence in the chloroplasts. The scale bar is 10 μm. (This figure is available in colour at JXB online.) mutant, we examined the levels of the RPS14 protein. The effects on ribosome levels, we examined the protein amounts level in the ECD1-RNAi transgenic plants decreased to less for two other plastid-encoded ribosomal proteins, RPS3 and than one-quarter of that in the wild-type in the cotyledons, RPL2. by immunoblot analysis. Both proteins were decreased but there was no change in the true leaves (Fig. 9A). RPS14 significantly in the ECD1-RNAi-1 line (Fig.  9B). Thus, the is required for the accumulation of ribosomal 30S subunits. To defect in the accumulation of RPS14 protein in the ECD1- determine whether the reduced content of RPS14 had any RNAi transgenic plants may have compromised ribosome Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 3046 | Jiang et al. Fig. 7. Gel electrophoretic mobility shift assays (EMSAs) with the rps14 oligoribonucleotides. (A) The sequences used as oligonucleotide probes. Asterisks indicate the editing sites. (B) EMSA showing that MBP-ECD1 binds to the sequence around rps14-149, but not rps14-80. Increasing concentrations of protein (25, 50 and 100 nM) were incubated with 10 nM probes. The positions of the shift and free probes are indicated. (C) Unlabeled oligoduplexes with 10-fold, 50-fold, and 100-fold excess were used for competition to confirm the specific interaction between ECD1 and the cis- element around rps14-149. (This figure is available in colour at JXB online.) Fig. 8. RNA immunoprecipitation analysis of ECD1 and rps14-149. (A) ECD1-FLAG protein accumulation in the 35S::ECD1-FLAG transgenic seedlings compared with the wild-type (WT). (B) RNA immunoprecipitation analysis. IP+, anti-FLAG immunoprecipitation; IP-, mock immunoprecipitation. rps14-80, rps14-149, and petL are fragments that contain the editing sites of rps14-80, rps14-149, and petL, respectively. Data are means (±SE) obtained from three replicates. accumulation, and the deficiency in translation was presumably unchanged, while Class  II genes (transcribed by both NEP responsible for the albino phenotype of cotyledons. and PEP) were differentially regulated in ECD1-RNAi-1. The transcript abundance of other chloroplast genes that are not clearly classified are shown in Supplementary Fig. S8. In order The ECD1 mutation affects plastid gene expression to verify these results, we carried out RNA blot analysis of the and plastid rRNA processing psbA, rbcL, clpP, and rpoA genes using sequence-specific labeled The expression of chloroplast genes significantly impacts on probes. The steady-state levels of transcripts were in almost chloroplast development. We examined the transcript abun- complete agreement with the qRT-PCR analysis, and the pro- dance of various chloroplast genes in 7-d-old seedlings by cessing patterns between the wild-type and ECD1-RNAi-1 qRT-PCR. The results showed that the transcript levels of plants did not differ greatly (Fig. 10B , Supplementary Fig. S9). Class I genes (transcribed preferentially by PEP) were signifi- These results indicated that ECD1 is essential for PEP but not cantly reduced in ECD1-RNAi-1 compared with the wild- for NEP activity. type (Fig. 10A). In contrast, transcript levels of Class III genes Decreased levels of rRNAs were found in the cotyle- (transcribed preferentially by NEP) were either increased or dons of the ECD1-RNAi transgenic plants using ethidium Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 RNA-editing factor ECD1 and chloroplast development | 3047 Fig. 9. Immunoblot of chloroplast ribosomal subunits in the wild-type (WT) and ECD1-RNAi-1 line. (A) Immunoblots of PRS14 in 7- and 14-d-old seedlings. C and T refer to the proteins in cotyledons and true leaves, respectively, in the 14-d samples. (B) Immunoblot of RPS3 and RPL2 in samples of whole seedlings at 7 d old. bromide-stained agarose gel assays (Fig.  10B, Supplementary especially embryo-lethal mutants. Chloroplast development is Fig.  S9). We next examined the role of ECD1 in rRNA closely related to embryo development and, as a consequence, metabolism. Chloroplast ribosomal RNAs are co-transcribed eliminating biosynthetic functions within the chloroplast and as a single RNA precursor that contains 16S, 23S, 4.5S, and 5S interfering with expression of the chloroplast genome often rRNAs, as well as two tRNAs (Fig. 11A). The precursor tran- results in embryo lethality in Arabidopsis (Bryant et al., 2011). script undergoes a complex series of processing events before A  set of 119 nuclear genes encoding chloroplast-localized maturation. The 23S–4.5S bi-cistronic RNA (3.2 kb) under- proteins has been identified, including many PPR proteins goes endonucleolytic cleavage to produce a mature 4.5S rRNA (Bryant et  al., 2011). Disruption of these genes results in an and a 23S precursor (2.9 kb), which undergoes further matu- embryo-defective phenotype, highlighting the importance ration and ultimately generates three species of 1.1, 1.3, and of chloroplasts in embryogenesis. Here, we identified a novel 0.5 kb. The 16S precursor RNA (1.7 kb) is processed to a 1.5- PPR protein, ECD1, which is required for chloroplast devel- kb mature 16S rRNA. RNA gel blot analysis was performed opment in seedlings. A  complete deletion of ECD1 resulted on the total leaf RNAs from 7- and 14-d-old wild-type and in embryo lethality, indicating that ECD1 is indispensable for ECD1-RNAi-1 plants using rRNA-specific probes (indicated plant growth and development. Evidence suggests that accD, in Fig.  11). Higher transcript levels of the 3.2-kb 23S–4.5S which encodes one subunit of a multimeric acetyl-CoA carb- rRNA precursor and the 1.7-kb 16S rRNA precursor were oxylase required for fatty acid biosynthesis, is amongst the most detected in the cotyledons of ECD1-RNAi-1, whereas the lev- important chloroplast genes required for embryo development els of the 1.5-kb 16S, 0.5-kb 23S, 0.1-kb 4.5S, and 0.12-kb in Arabidopsis (Bryant et al., 2011; Parker et al., 2016). Editing 5S mature rRNAs decreased drastically (Fig.  11B). However, of accD-794 and accD-58642 both decreased in the ECD1- no obvious differences in the levels of the rRNA transcripts RNAi transgenic plants (Supplementary Fig. S4). However, the between the wild-type and ECD1-RNAi-1 were observed in editing deficiency of these sites may have been caused indir- true leaves (Supplementary Fig. S10). On the other hand, the ectly by the albino phenotype since they were also affected in transcript levels of the two tRNAs encoded by this operon, trnI seedlings treated with lincomycin or spectinomycin. and trnA, accumulated to the same extent in ECD1-RNAi-1 In the ECD1-RNAi transgenic plants, the chloroplast ultra- and the wild-type cotyledons (Supplementary Fig. S11). These structure in cotyledons exhibited abnormal morphology and results suggested that ECD1 may play an important role in the thylakoid membrane structure was perturbed, suggesting plastid rRNA maturation. that the albino phenotype of cotyledons in these plants was probably due to developmentally defective chloroplasts. Proper accumulation of plastid ribosomal proteins is a prerequisite for assembling functional ribosomes and is necessary for chloro- Discussion plast development. A deficiency of nuclear-encoded chloroplast Chloroplasts are organelles that perform photosynthesis. The factors required for the synthesis of plastid ribosomal proteins development of a functional chloroplast is regulated by a large can compromise assembly and accumulation of chloroplast number of genetic factors, especially nuclear-encoded factors. ribosomes. PPR4 is required for the trans-splicing of the plas- Given their importance, extensive studies have been carried tid rps12 transcript and consequently affects the accumulation out focusing on the identification of nuclear genes essential of plastid ribosomes (Schmitz-Linneweber et  al., 2006). The for chloroplast development. However, numerous nuclear maize chloroplast protein PPR103 stabilizes the 5´-end of pro- mutants that impact on chloroplast development can easily cessed rpl16 mRNAs and a loss of plastid ribosomes was also be overlooked because of their severe, often lethal phenotype, detected in ppr103 mutants (Hammani et al., 2016). Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 3048 | Jiang et al. Fig. 10. Chloroplast gene expression in ECD1-RNAi-1 relative to the wild-type (WT). (A) Transcript levels of chloroplast genes (Classes I–III) were measured by quantitative RT-PCR (qRT-PCR). Data are given as log of ECD1-RNAi-1/WT ratios from at least three independent experiments. Class I genes refer to psbC, psbT, psbN, psbH, psbD, psbB, petD, psbA, ndhA, psbE, psbF, and rbcL. Class II genes refer to rrn23, rrn16, atpI, ycf1, clpP, rps16, ndhB, ndhF, atpE, psaJ, and rps18. Class III genes refer to accD, ycf2, rpoB, rpoC2, and rpoC1. (B) RNA blot analysis of transcript levels for the different chloroplast gene classes. An ethidium bromide-staining gel is shown as a loading control (EtBR). In cotyledons of the ECD1-RNAi transgenic plants, the complexes with other proteins (Hammani et al., 2009). RNA- accumulation of RPS14 decreased dramatically compared editing of rps14-149 changes Pro to Leu. Since Pro tends to with the wild-type, which may have been due to the edit- disrupt α-helices and thus leads to instability of proteins, edit- ing defects in rps14-149. We found that ECD1 is able to bind ing of this site could restore an α-helix and stabilize the RPS14 to the cis-element of rps14-149 but not to the other editing protein (Sugita et al., 2006). RPS14 is essential for the assem- site rps14-80, indicating that ECD1 specifically affects RNA- bly of the ribosomal 30S subunit and contributes to the pep- editing of rps14-149. RNA-editing defects result in amino tide environment of the peptidyl transferase center in E.  coli acid changes that may directly alter protein function, or act (Brochier et al., 2000), and is also essential in tobacco plastids by destabilizing the protein or by affecting its ability to form (Ahlert et  al., 2003; Tiller and Bock, 2014). In Physcomitrella Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 RNA-editing factor ECD1 and chloroplast development | 3049 Fig. 11. Expression and processing of chloroplast rRNA. (A) Schematic representation of the chloroplast rrn operon of Arabidopsis. The locations of probes (a–j) used for the RNA gel blot analysis and the size of transcripts (in kb) are indicated below the operon. (B) RNA gel blot analysis. Mature transcripts were decreased and the precursors were increased in ECD1-RNAi-1 compared to the wild-type (WT). RNA was extracted from 7-d-old seedlings. An ethidium bromide-staining gel is shown as a loading control (EtBr). patens, reduction of RNA editing in rps14-C2 impairs the specific defect in 16S rRNA maturation in a cotyledon-spe- translation of the RPS14 protein and affects the function of cific manner (Yamamoto et al., 2000). sco1 mutants are defec- the chloroplast ribosome, which then results in a pale-green tive in the chloroplast elongation factor G, which not only phenotype and decreased photosynthetic activity in PpPPR- affects chloroplast mRNA translation during chloroplast for- 45-RNAi plants (Ichinose et  al., 2014). Thus, the decrease of mation in cotyledons, but also other developmental processes RPS14 may block the proper assembly of plastid ribosomes such as germination and flowering (Albrecht et al., 2006). The in Arabidopsis. A deficiency in plastid ribosomes may account proteins affected by sco2/cyo1 are required for protein folding for the global defects in PEP-dependent transcripts, as transla- and both of them have DnaJ-like zinc finger domains (Shimada tion of core subunits of the plastid-encoded RNA polymer- et al., 2007; Albrecht et al., 2008). The expression of ECD1 is ase decreases to a level that is not sufficient for transcribing not limited to cotyledons only; however, the higher amount in the PEP-dependent genes (Hess et al., 1993; Zubko and Day, cotyledons and its specific role in RNA-editing of rps14-149 2002; Chi et al., 2008). As ribosome assembly and pre-rRNA in cotyledons but not in true leaves suggest that ECD1 is more processing are intimately linked, defects in rRNA processing important in cotyledons than in leaves and other organs. Since in the cotyledons of ECD1-RNAi-1 may also be the con- ECD1-RNAi-1 is not lethal, the editing of rps14-149 is not sequence of the deficiency in ribosome assembly (Charollais completely abolished. It is expected that more severe editing et  al., 2003; Granneman and Baserga, 2005). In addition, the defects occur in the ecd1 homozygotes, leading to an embryo- decreased accumulation of RPL2 and RPS3 provides fur- lethal phenotype. ther evidence for a deficiency of chloroplast ribosomes. Taken In conclusion, the results of our study indicate that the PPR together, insufficient accumulation of plastid ribosomes may protein ECD1 is a site-specific factor for the rps14-149 RNA- be the cause for the developmentally defective chloroplasts in editing site, and it is required for early chloroplast development cotyledons of ECD1-RNAi-1. in seedlings. A decrease in ECD1 expression leads to editing In the ECD1-RNAi transgenic plants, chloroplast develop- defects of rps14-149 in cotyledons, which result in decreased ment within cotyledons (but not in true leaves) was severely accumulation of the RPS14 protein; this in turn leads to impaired, leading to the formation of white cotyledons. In lower levels of plastid ribosomes in cotyledons, and thus to recent years, several mutants with a phenotype of cotyledon- defects in chloroplast development. Decreased expression of specific impairment have been isolated. wco mutants have a PEP-dependent genes and defective plastid rRNA processing Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 3050 | Jiang et al. Brochier C, Philippe H, Moreira D. 2000. The evolutionary history were also observed in ECD1-RNAi transgenic plants, which of ribosomal protein RpS14: horizontal gene transfer at the heart of the can largely be accounted for by limiting amounts of plastid ribosome. Trends in Genetics 16, 529–533. ribosomes. Bryant N, Lloyd J, Sweeney C, Myouga F, Meinke D. 2011. Identification of nuclear genes encoding chloroplast-localized proteins required for embryo development in Arabidopsis. Plant Physiology 155, 1678–1689. Cai W, Ji D, Peng L, Guo J, Ma J, Zou M, Lu C, Zhang L. 2009. LPA66 Supplementary data is required for editing psbF chloroplast transcripts in Arabidopsis. Plant Physiology 150, 1260–1271. Supplementary data are available at JXB online. Charollais J, Pflieger D, Vinh J, Dreyfus M, Iost I. 2003. The DEAD-box Fig. S1. Motifs and sequence analysis of ECD1. RNA helicase SrmB is involved in the assembly of 50S ribosomal subunits Fig. S2. Expression patterns of ECD1 within seedlings. in Escherichia coli. Molecular Microbiology 48, 1253–1265. Fig.  S3. Ultrastructure of chloroplasts in ECD1-RNAi-1 Chi W, He B, Manavski N, Mao J, Ji D, Lu C, Rochaix JD, Meurer and wild-type seedlings. J, Zhang L. 2014. RHON1 mediates a Rho-like activity for transcription termination in plastids of Arabidopsis thaliana. The Plant Cell 26, 4918–4932. Fig. S4. RNA-editing efficiency of var ious target sites in plas- Chi W, Ma J, Zhang D, Guo J, Chen F, Lu C, Zhang L. 2008. The tids of cotyledons in ECD1-RNAi-1 and wild-type seedlings, pentratricopeptide repeat protein DELAYED GREENING1 is involved in the and in seedlings treated with lincomycin or spectinomycin. regulation of early chloroplast development and chloroplast gene expression Fig. S5. RNA editing efficiency of rps14 transcripts in seed- in Arabidopsis. Plant Physiology 147, 573–584. lings treated with lincomycin or spectinomycin. Clough SJ, Bent AF. 1998. Floral dip: a simplified method for Fig.  S6. Plastid RNA editing in true leaves of ECD1- Agrobacterium-mediated transformation of Arabidopsis thaliana. The Plant Journal 16, 735–743. RNAi-1 and wild-type seedlings. Cushing DA, Forsthoefel NR, Gestaut DR, Vernon DM. 2005. Fig. S7. Purification of MBP-ECD1. Arabidopsis emb175 and other ppr knockout mutants reveal essential roles Fig. S8. Transcript levels of chloroplast genes for which the for pentatricopeptide repeat (PPR) proteins in plant embryogenesis. Planta transcribing RNA polymerase is unknown. 221, 424–436. Fig.  S9. RNA gel blot analysis of chloroplast RNAs in Granneman S, Baserga SJ. 2005. Crosstalk in gene expression: coupling and co-regulation of rDNA transcription, pre-ribosome assembly and pre- cotyledons and true leaves of 14-d-old wild-type and ECD1- rRNA processing. Current Opinion in Cell Biology 17, 281–286. RNAi-1 seedlings. Hammani K, Okuda K, Tanz SK, Chateigner-Boutin AL, Shikanai T, Fig.  S10. RNA gel blot analysis of rRNAs in cotyledons Small I. 2009. A study of new Arabidopsis chloroplast RNA editing mutants and true leaves of 14-d-old wild-type and ECD1-RNAi-1 reveals general features of editing factors and their target sites. The Plant Cell 21, 3686–3699. seedlings. Hammani K, Takenaka M, Miranda R, Barkan A. 2016. A PPR protein in Fig. S11. RNA gel blot analysis of trnI and trnA in cotyledons the PLS subfamily stabilizes the 5´-end of processed rpl16 mRNAs in maize and true leaves of wild-type and ECD1-RNAi-1 seedlings. chloroplasts. Nucleic Acids Research 44, 4278–4288. Table S1. List of primers used in this study. Hashimoto M, Endo T, Peltier G, Tasaka M, Shikanai T. 2003. A nucleus-encoded factor, CRR2, is essential for the expression of chloroplast ndhB in Arabidopsis. The Plant Journal 36, 541–549. Acknowledgements Hedtke B, Börner T, Weihe A. 1997. Mitochondrial and chloroplast phage-type RNA polymerases in Arabidopsis. Science 277, 809–811. We are grateful to Prof. Jean David Rochaix for revising the manu- Hess WR, Prombona A, Fieder B, Subramanian AR, Börner T. 1993. script. We thank Tiegang Lu (The Chinese Academy of Agricultural Chloroplast rps15 and the rpoB/C1/C2 gene cluster are strongly transcribed Sciences) for providing us with antibodies for the plastid-encoded ribo- in ribosome-deficient plastids: evidence for a functioning non-chloroplast- somal proteins RPS3 and RPL2. This study was supported by the State encoded RNA polymerase. The EMBO Journal 12, 563–571. Key Basic Research and Development Plan of China (2015CB150100) Ichinose M, Uchida M, Sugita M. 2014. Identification of a pentatricopeptide and the Youth Innovation Promotion Association CAS (2013059). repeat RNA editing factor in Physcomitrella patens chloroplasts. FEBS Letters 588, 4060–4064. Kim BH, Malec P, Waloszek A, von Arnim AG. 2012. Arabidopsis BPG2: References a phytochrome-regulated gene whose protein product binds to plastid ribosomal RNAs. Planta 236, 677–690. Ahlert D, Ruf S, Bock R. 2003. Plastid protein synthesis is required for plant development in tobacco. Proceedings of the National Academy of Kovtun Y, Chiu WL, Tena G, Sheen J. 2000. Functional analysis of Sciences, USA 100, 15730–15735. oxidative stress-activated mitogen-activated protein kinase cascade in plants. Proceedings of the National Academy of Sciences, USA 97, Albrecht V, Ingenfeld A, Apel K. 2006. Characterization of the snowy 2940–2945. cotyledon 1 mutant of Arabidopsis thaliana: the impact of chloroplast elongation factor G on chloroplast development and plant vitality. Plant Liere K, Maliga P. 2001. Plastid RNA polymerases in higher plants. In: Aro Molecular Biology 60, 507–518. E-M, Andersson B, eds. Regulation of photosynthesis. Dordrecht: Springer Netherlands, 29–49. Albrecht V, Ingenfeld A, Apel K. 2008. Snowy cotyledon 2: the identification of a zinc finger domain protein essential for chloroplast development in Lurin C, Andrés C, Aubourg S, et al. 2004. Genome-wide analysis of cotyledons but not in true leaves. Plant Molecular Biology 66, 599–608. Arabidopsis pentatricopeptide repeat proteins reveals their essential role in organelle biogenesis. The Plant Cell 16, 2089–2103. Aryamanesh N, Ruwe H, Sanglard LV, Eshraghi L, Bussell JD, Howell KA, Small I, des Francs-Small CC. 2017. The pentatricopeptide repeat Martínez-García JF, Monte E, Quail PH. 1999. A simple, rapid and protein EMB2654 is essential for trans-splicing of a chloroplast small quantitative method for preparing Arabidopsis protein extracts for ribosomal subunit transcript. Plant Physiology 173, 1164–1176. immunoblot analysis. The Plant Journal 20, 251–257. Barkan A, Small I. 2014. Pentatricopeptide repeat proteins in plants. Meinke DW. 1994. Seed development in Arabidopsis thaliana. In: Annual Review of Plant Biology 65, 415–442. Meyerowitz EM, Somerville CR, eds. Arabidopsis. New York: Cold Spring Harbor Laboratory Press, 253–259. Börner T, Aleynikova AY, Zubo YO, Kusnetsov VV. 2015. Chloroplast RNA polymerases: role in chloroplast biogenesis. Biochimica et Biophysica Mullet JE. 1993. Dynamic regulation of chloroplast transcription. Plant Acta 1847, 761–769. Physiology 103, 309–313. Downloaded from https://academic.oup.com/jxb/article/69/12/3037/4965867 by DeepDyve user on 19 July 2022 RNA-editing factor ECD1 and chloroplast development | 3051 Parker N, Wang Y, Meinke D. 2016. Analysis of Arabidopsis accessions Sugita M, Miyata Y, Maruyama K, Sugiura C, Arikawa T, Higuchi M. hypersensitive to a loss of chloroplast translation. Plant Physiology 172, 2006. Extensive RNA editing in transcripts from the PsbB operon and RpoA 1862–1875. gene of plastids from the enigmatic moss Takakia lepidozioides. Bioscience, Biotechnology, and Biochemistry 70, 2268–2274. Pfalz J, Bayraktar OA, Prikryl J, Barkan A. 2009. Site-specific binding of a PPR protein defines and stabilizes 5´ and 3´ mRNA termini in chloroplasts. Sun X, Feng P, Xu X, Guo H, Ma J, Chi W, Lin R, Lu C, Zhang L. 2011. A The EMBO Journal 28, 2042–2052. chloroplast envelope-bound PHD transcription factor mediates chloroplast signals to the nucleus. Nature Communications 2, 477. Pryor KD, Leiting B. 1997. High-level expression of soluble protein in Escherichia coli using a His6-tag and maltose-binding-protein double- Tiller N, Bock R. 2014. The translational apparatus of plastids and its role affinity fusion system. Protein Expression and Purification 10, 309–319. in plant development. Molecular Plant 7, 1105–1120. Ruwe H, Castandet B, Schmitz-Linneweber C, Stern DB. 2013. Tseng CC, Lee CJ, Chung YT, Sung TY, Hsieh MH. 2013. Differential Arabidopsis chloroplast quantitative editotype. FEBS Letters 587, regulation of Arabidopsis plastid gene expression and RNA editing in non- 1429–1433. photosynthetic tissues. Plant Molecular Biology 82, 375–392. Sakamoto W, Miyagishima SY, Jarvis P. 2008. Chloroplast biogenesis: Tzafrir I, Pena-Muralla R, Dickerman A, et  al. 2004. Identification of control of plastid development, protein import, division and inheritance. The genes required for embryo development in Arabidopsis. Plant Physiology Arabidopsis Book 6, e0110. 135, 1206–1220. Sato S, Nakamura Y, Kaneko T, Asamizu E, Tabata S. 1999. Complete Walter M, Chaban C, Schütze K, et  al. 2004. Visualization of structure of the chloroplast genome of Arabidopsis thaliana. DNA Research protein interactions in living plant cells using bimolecular fluorescence 6, 283–290. complementation. The Plant Journal 40, 428–438. Schmitz-Linneweber C, Small I. 2008. Pentatricopeptide repeat proteins: a Wang Y, Ren Y, Zhou K, et al. 2017. WHITE STRIPE LEAF4 encodes a socket set for organelle gene expression. Trends in Plant Science 13, 663–670. novel P-type PPR protein required for chloroplast biogenesis during early leaf development. Frontiers in Plant Science 8, 1116. Schmitz-Linneweber C, Williams-Carrier RE, Williams-Voelker PM, Kroeger TS, Vichas A, Barkan A. 2006. A pentatricopeptide repeat Yamamoto YY, Puente P, Deng XW. 2000. An Arabidopsis cotyledon- protein facilitates the trans-splicing of the maize chloroplast rps12 pre- specific albino locus: a possible role in 16S rRNA maturation. Plant & Cell mRNA. The Plant Cell 18, 2650–2663. Physiology 41, 68–76. Shikanai T, Fujii S. 2013. Function of PPR proteins in plastid gene Zhou W, Lu Q, Li Q, Wang L, Ding S, Zhang A, Wen X, Zhang L, Lu C. expression. RNA Biology 10, 1446–1456. 2017. PPR-SMR protein SOT1 has RNA endonuclease activity. Proceedings Shimada H, Mochizuki M, Ogura K, et al. 2007. Arabidopsis cotyledon- of the National Academy of Sciences, USA 114, E1554–E1563. specific chloroplast biogenesis factor CYO1 is a protein disulfide isomerase. Zubko MK, Day A. 2002. Differential regulation of genes transcribed by The Plant Cell 19, 3157–3169. nucleus-encoded plastid RNA polymerase, and DNA amplification, within Small ID, Peeters N. 2000. The PPR motif – a TPR-related motif prevalent ribosome-deficient plastids in stable phenocopies of cereal albino mutants. in plant organellar proteins. Trends in Biochemical Sciences 25, 46–47. Molecular Genetics and Genomics 267, 27–37.

Journal

Journal of Experimental BotanyOxford University Press

Published: May 25, 2018

There are no references for this article.