AtECB2, a pentatricopeptide repeat protein, is required for chloroplast transcript accD RNA editing and early chloroplast biogenesis in Arabidopsis thaliana


  • Qing-Bo Yu,

    1. Key Laboratory of Photosynthesis and Environmental Molecular Physiology, Institute of Botany, Chinese Academy of Sciences, Beijing 100093, China
    2. Biology Department, Life and Environmental College, Shanghai Normal University, Shanghai 200234, China
    3. Graduate School of the Chinese Academy of Sciences, Beijing 100049, China
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Yan Jiang,

    1. Biology Department, Life and Environmental College, Shanghai Normal University, Shanghai 200234, China
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Kang Chong,

    Corresponding author
    1. Key Laboratory of Photosynthesis and Environmental Molecular Physiology, Institute of Botany, Chinese Academy of Sciences, Beijing 100093, China
    Search for more papers by this author
  • Zhong-Nan Yang

    Corresponding author
    1. Biology Department, Life and Environmental College, Shanghai Normal University, Shanghai 200234, China
    Search for more papers by this author


This article is corrected by:

  1. Errata: Correction Volume 61, Issue 3, 544, Article first published online: 21 January 2010

*(fax +86 21 6432 4190; e-mail and fax +86 10 8259 4821; e-mail


Chloroplast biogenesis is a complex process in higher plants. Screening chloroplast biogenesis mutants, and elucidating their molecular mechanisms, will provide insight into the process of chloroplast biogenesis. In this paper, we obtained an early chloroplast biogenesis mutant atecb2 that displayed albino cotyledons and was seedling lethal. Microscopy observations revealed that the chloroplast of atecb2 mutants lacked an organized thylakoid membrane. The AtECB2 gene, which is highly expressed in cotyledons and seedlings, encodes a pentatricopeptide repeat protein (PPR) with a C-terminal DYW domain. The AtECB2 protein is localized in the chloroplast, and contains a conserved HxExnCxxC motif that is similar to the activated site of cytidine deaminase. The AtECB2 mutation affects the expression pattern of plastid-encoded genes. Immunoblot analyses showed that the levels of photosynthetic proteins decreased substantially in atecb2 mutants. Inspection of all reported plastid RNA editing sites revealed that one editing site, accD, is not edited in atecb2 mutants. Therefore, the AtECB2 protein must regulate the RNA editing of this site, and the dysfunctional AccD protein from the unedited RNA molecules could lead to the mutated phenotype. All of these results indicate that AtECB2 is required for chloroplast transcript accD RNA editing and early chloroplast biogenesis in Arabidopsis thaliana.


In higher plants, chloroplasts develop from proplastids. Proplastids are small, round organelles with very few internal membranes that are found primarily in meristematic cells, as well as in young postmitotic cells. During the process of chloroplast biogenesis, proplastids are first enlarged, and then, in the presence of light, the inner membrane of the proplastid invaginates into vesicles. Subsequently, these vesicles fuse into a lamellar structure that is later complemented by smaller, disc-shaped structures that develop into grana stacks. At the same time, the developing chloroplast forms into its typical lens shape. Finally, the mature chloroplast is formed, containing an intertwined internal membrane system: the thylakoid membrane (Vothknecht and Westhoff, 2001; Gutierrez-Nava et al., 2004; Pyke, 2007). Evidently, chloroplast biogenesis is a complex process. Previous reports have identified a significant proportion of nuclear genes from various chloroplast biogenesis mutants. These genes are involved in processes such as RNA processing and editing (Fisk et al., 1999; Chateigner-Boutin et al., 2008; de Longevialle et al., 2008), protein translation and folding (Sundberg et al., 1997; Bauer et al., 2000; Motohashi et al., 2001) and plastidic isoprenoid biosynthesis (Mandel et al., 1996; Gutierrez-Nava et al., 2004; Guevara-Garcia et al., 2005). Several of these genes belong to the pentatricopeptide repeat protein (PPR) family, which is one of the largest gene families in plants.

The PPR family is defined by the signature motif of a degenerate 35-amino-acid repeat, often arranged in tandem arrays of 2–27 repeats per peptide (Small and Peeters, 2000). The family members are divided into two subfamilies: the P subfamily, with members composed of PPR motifs only, and the PLS subfamily, with repeated blocks of P-L-S motifs, also referred to as the plant combinatorial and modular protein family (PCMPs) (Aubourg et al., 2000; Lurin et al., 2004). The majority of the PPR proteins are predictably targeted to either the mitochondria or the chloroplast. In the Arabidopsis thaliana genome, the PPR family contains 466 members, and roughly half of the PPR proteins are in the PLS subfamily, which is specific to land plants. The PLS subfamily is further subdivided into four subgroups, including PLS, E, E+ and DYW, according to the variable tandem repeat non-PPR motifs at the C terminus (Lurin et al., 2004). Many studies have shown that the PPR genes influence numerous biological processes, including RNA splicing and translation (Hashimoto et al., 2003; Meierhoff et al., 2003; Nakamura et al., 2003; Schmitz-Linneweber et al., 2006; Uyttewaal et al., 2008), RNA editing (Kotera et al., 2005; Okuda et al., 2007, 2009; Chateigner-Boutin et al., 2008; Zhou et al., 2008; Zehrmann et al., 2009), RNA stability (Yamazaki et al., 2004) and ribosome accumulation (Williams and Barkan, 2003). Additionally, PPR proteins are involved in embryogenesis (Cushing et al., 2005), and in restoring the fertility of cytoplasmic male sterility (Wang et al., 2006). However, the molecular functions of only a few PPR proteins have been characterized in detail, and there is still a great deal of work to be done to determine the functions of the other PPR proteins in plant development.

RNA editing is a post-transcriptional process that alters an RNA sequence from that encoded by the genome. RNA editing events have been reported in a wide range of organisms, from viruses to mammals and plants (Kotera et al., 2005; Shikanai, 2006). In many editing events, especially in plant plastid organelles, RNA editing is essential for expressing functional proteins by modifying amino acid sequences or generating a translational start codon (Hoch et al., 1991; Kotera et al., 2005) or a stop codon (Wintz et al., 1991). The number of site-specific editing sites varies considerably between organisms. In A. thaliana, 34 editing sites have been detected in the genome of plastids (Tillich et al., 2005; Chateigner-Boutin and Small, 2007). So far, only six PPR proteins responsible for plastid-specific RNA editing events have been isolated from A. thaliana. CRR4 is required for editing at the ndhD-1 site, which creates the initial translational codon of the plastid ndhD gene (Kotera et al., 2005). CRR21 is involved in another editing site ndhD-2 for the gene, and the editing event convert Ser-128 of NdhD to Leu (Okuda et al., 2007). CLB19 is essential for editing the two sites of the plastid gene rpoA (the 67th codon of rpoA transcript) and clpP transcripts (the 187th codon of clpP1 transcript) (Chateigner-Boutin et al., 2008). YS1 participates in the editing of the 25 992 site in the rpoB transcript, and affects the rapid development of chloroplasts during early growth (Zhou et al., 2008). More recently, another two PPR proteins, CRR22 and CRR28, were reported to be involved in several RNA-editing events in chloroplasts (Okuda et al., 2009). These studies only identified a few specific factors for the RNA editing of chloroplast transcripts. Therefore, it is important to identify other factors for RNA-editing events in the chloroplast.

In this paper, we reported on the characterization of a PPR family gene, AtECB2, which showed a high level of expression in cotyledons and seedlings, and analyzed atecb2 mutants in detail. AtECB2, a member of the DYW subgroup, regulated the RNA editing of the chloroplast transcript accD, and is essential for the early stages of chloroplast biogenesis.


Isolation of the atecb2 mutant and its phenotypic characterization

To identify candidate genes essential for early chloroplast biogenesis, we obtained the knock-out transgenic lines for 34 genes, the products of which are chloroplast targeted and are highly expressed in early-stage seedlings. One T-DNA insertion line (SALK_112251) of the gene AT1G15510, designated atecb2, displayed a low accumulation of photosynthetic pigments, and thus it was further investigated in detail. The atecb2 mutant showed an albino cotyledon without a primary leaf, and only survived for roughly 1 week when grown in a plant nutrient solution (PNS) medium (Haughn and Somerville, 1986; Figure 1a,b). Therefore, it was unable to grow photoautotrophically, and was seedling lethal under autotrophic growth conditions. When supplemented with sucrose as a carbon source, the atecb2 mutant was able to produce only two albino primary leaves before it stopped growing (Figure 1c,d). Neither viability nor pigmentation of this mutant was restored under low-light growing conditions (60 μmol m−2 s−1; data not shown). Thus, this phenotype is not caused by photo-oxidative stress.

Figure 1.

 Phenotypes of atecb2 mutants and wild-type (WT) Arabidopsis (Columbia ecotype), grown on a plant nutrient solution (PNS) medium with or without 2% sucrose: (a) 7-day-old wild type and (b) atecb2 mutant grown on a PNS medium; (c) 14-day-old wild type and (d) atecb2 mutant grown on a PNS medium with 2% sucrose; (e) 7-day-old and (f) 14-day-old complementary plant (TC), respectively, grown on PNS medium with 2% sucrose. Scale bars: 2 mm.

We quantified the levels of chlorophyll a and b in both the mutant and the wild type. The total chlorophyll content of atecb2 mutants (110.7 ± 10.7 μg g−1 fresh weight for the atecb2 mutant type) was lower than that of the wild type (691.63 ± 4.3 μg g−1 fresh weight for the wild type). The photosynthetic activity of atecb2 mutants was also investigated. We analyzed the Fv/Fm = (Fm – F0)/Fm value, which is an indicator of photo-inhibition (F0 and Fm are the minimal and maximal chlorophyll a fluorescence of dark-adapted leaves, respectively; Fv indicates the maximum variable fluorescence Meurer et al., 1996). The Fv/Fm of the 14-day-old wild type was determined to be 0.81 ± 0.01 (n = 3), but the value could not be calculated in atecb2 mutants. The results indicated that photosynthetic activity was seriously damaged in atecb2 mutants.

As the atecb2 mutant showed an albino phenotype, we compared its chloroplast morphology and ultrastructure with that of the wild type by transmission electron microscopy (TEM). In the mature chloroplasts of the 14-day-old wild type, an organized thylakoid membrane was observed. The stroma thylakoid and grana thylakoid were distinguished in chloroplasts of the wild type (Figure 2a). Additionally, the chloroplast contained starch grains. Conversely, chloroplasts in atecb2 leaves were smaller than those in the wild type, and no organized thylakoid membranes were observed, except for a few short appressed internal membranes. Moreover, the plastids observed in atecb2 mutants did not contain starch grains (Figure 2b). These features indicate that the organelle was arrested at an early stage of chloroplast biogenesis.

Figure 2.

 Ultrastructure of chloroplasts from 14-day-old wild-type, atecb2 and complementary plant leaves.
Chloroplasts are from the first leaf of: (a) wild type; (b) atecb2 mutant; and (c) a 14-day-old complementary plant. Abbreviations: Cp, chloroplast; Sam, short appressed membrane; SG, Starch Grain; St, stroma; Thy, thylakoid. The chloroplasts were magnified 5000-fold. Scale bars: 10 μm.

Molecular complementation of the atecb2 mutation

Our genetic analysis showed that the albino phenotype of the atecb2 mutants was controlled by a recessive nuclear gene, and co-segregated with the kanamycin resistance marker (data not shown). The sequencing of the polymerase chain reaction (PCR) fragments showed that the T-DNA was inserted downstream of the 489th base pair (bp) from the start codon of the gene AT1G15510 (Figure 3a,b). Reverse-transcriptase PCR (RT-PCR) results indicated that the transcript of the gene was absent in atecb2 mutants, but present in the wild type (Figure 3c). All of these results showed that AT1G15510 was a candidate gene for AtECB2.

Figure 3.

 Insertion analysis of atecb2 mutants and the AtECB2 expression levels in wild-type (WT), atecb2 mutants and in complementary plants (TC).
(a) Schematic representation of the genomic region flanking the T-DNA insertions. The complementary fragment is shown, and the numbers represent the locus in the chromosome. The black box represents the translated region of AtECB2.
(b) Identification of atecb2 homozygotes, atecb2/+ heterozygotes and the wild type, using insertion site analysis.
(c) The AtECB2 expression levels in the wild type, atecb2 mutants and in the complementary plants.

To further confirm that the AtECB2 gene was AT1G15510, we performed a complement test with a genomic sequence. A 4876-bp wild-type genomic sequence, containing the AT1G15510 gene as well as its 1936-bp upstream and 339-bp downstream sequences, was introduced into the normal phenotype offspring from AtECB2/atecb2 heterozygotes via an Agrobacterium tumefaciens-mediated transformation (Clough and Bent, 1998). Then, 99 T1 transgenic plants were screened out. Among them, five plants that were AtECB2/atecb2 heterozygotes were obtained. The segregate rates of their offspring were far from 3:1 (green plants:albino plants; Table S1). Furthermore, we determined the genotype of these offspring, as described in the Experimental procedures, and obtained more than 10 atecb2/atecb2 homozygous plants from the sixth T1 transgenic lines that carried the fragments of the exogenous genomic sequence. These plants displayed green cotyledons and leaves after their germination (Figure 1e,f). RT-PCR analysis also showed that the transcripts of AT1G15510 were present in these plants (Figure 3c). Furthermore, TEM observations for these plants showed that the chloroplasts had similar ultrastructure to the wild-type plants (Figure 2c). These results indicate that the 4876-bp genomic fragments can successfully complement the mutated phenotype, and that the AtECB2 gene was AT1G15510.

The AtECB2 gene encodes a PPR protein localized in the chloroplast

BLAST searches of the complete Arabidopsis genomic sequence revealed that only one copy of the AtECB2 gene was present in the nuclear genome. To obtain the full-length AtECB2 cDNA sequence, we exploited the rapid amplification of cDNA ends (RACE), which resulted in the identification of the complete cDNA sequence. Sequence analysis indicated that the gene is not interrupted by introns, and it contains a 30-bp 5′-untranslated region (UTR), a 195-bp 3′-UTR and the entire coding sequence consistent with that in the Arabidopsis information resource database (TAIR) (Figure S1). The AtECB2 gene encodes a putative polypeptide of 866 amino acids, with a calculated molecular weight of 97.7 kDa. The protein is a member of the PPR family, and contains eleven characteristic PPR motifs, which can predictably form into a pair of antiparallel α-helices (Figure 4b). Besides these PPR motifs, this protein contains three extra conserved motifs E, E+ and DYW in the C terminal (Figure 4c). Thus, AtECB2 belongs to the DYW group of the PLS subfamily, as proposed by Lurin et al. (2004). Moreover, AtECB2 contains an unknown motif consisting of 15 amino acids (xGCSxI/VEExI/VExxGxV/IHxF), which is highly conserved in some PPR proteins, including CRR2, CLB19, CRR4, CRR21, YS1, CRR22 and CRR28. Importantly, this protein contains another conserved motif HxExnCxxC that is similar to the activated site of cytidine deaminases. This motif was also observed in the other four DYW proteins. The AtECB2 homologous proteins are conserved in grape, rice and maize (Figure 4a), and were very similar to the Vitis vinifera protein CAO46020 (81% similarity), the Zea maize protein AZM5_26406 (81% similarity) and the Oryza sativa protein Os05g0574800 (71% similarity).

Figure 4.

 Sequence analysis of AtECB2.
(a) Alignment analysis between AtECB2 and its homolog from Vitis vinifera, Zea maize and Oryza sativa. The sequences were aligned using ClustalW (Thompson et al., 1994). The arrowhead denotes the putative cleavage site of the transpeptide in AtECB2.
(b) Alignment analysis of the eleven motifs in the AtECB2 protein.
(c) Comparative analysis of three additional motifs, E, E+ and DYW, among CRR4, CRR21, CRR2, CLB19, YS1, CRR22 and CRR28. The conserved HxExnCxxC motif is also shown. The conserved 15-amino-acid (xGCSxI/VEExI/VExxGxV/IHxF) motif is marked.

The TargetP (Emanuelsson et al., 2000) and Predotar (Small et al., 2004) programs predicted that the AtECB2 protein is targeted to the chloroplast, and that its 52-amino-acid N-terminal region is likely to be a transit peptide. To confirm its subcellular localization, an AtECB2 N-terminal transit peptide fused with GFP driven by the CaMV 35S promoter was introduced into the wild-type plants. Stable transgenic plants were then examined with confocal laser microscopy, and the GFP fusion protein with the AtECB2 N-terminal transit peptides was co-localized with chloroplast autofluorescence (Figure 5g–i). With just the control vector, GFP fluorescence accumulated in the cytoplasm (Figure 5d–f), whereas no signals were found in the wild type (Figure 5a–c). Therefore, it appears that the 52-amino-acid sequence of the AtECB2 N-terminal region functions as a transit peptide, and AtECB2 is localized in the chloroplast.

Figure 5.

 Cellular localization of AtECB2 by GFP assays.
Fluorescence signals were visualized using confocal laser scanning microscopy. Green fluorescence indicates GFP, red fluorescence shows chloroplast autofluorescence and orange/yellow fluorescence shows merged images of the two types of fluorescence. Chlorophyll autofluorescence was detected (a, d and g). Fluorescence from GFP detected in Arabidopsis mesophyll cells (b, e and h). Scale bars: 10 μm.

The AtECB2 gene expression pattern

Expression data from Genvestigator showed that AtECB2 was widely expressed in Arabidopsis, but the highest expressional level was observed in seedlings (Zimmermann et al., 2004; RT-PCR analysis revealed that the expression abundance of the AtECB2 gene was highest in 14-day-old seedlings, compared with that in other tissues (Figure 6a). A quantitative real-time RT-PCR analysis confirmed this expression pattern of the gene, which is in accordance with that in the microarray data (Figure 6b).

Figure 6.

AtECB2 expression pattern in different tissues of Arabidopsis.
(a) Semiquantitative RT-PCR analysis and (b) quantitative real-time RT-PCR analysis of the AtECB2 gene in the root (Rt), stem (St), leaf (Le), flower (Fl) and seedling (Se).
(c) β-Glucuronidase (GUS) activity was observed in germinating seeds and (d) in 3-day-old cotyledons.
(e) GUS staining was observed in the stem and cauline leaf, and (f) in 14-day-old seedlings.
(g) GUS activity was observed in the sepals, stamens and styles, but not in the petals.
(h) GUS staining was observed in maturing siliques, but not in mature seeds (i).

The expression pattern of AtECB2 in Arabidopsis was further investigated using a β-glucuronidase (GUS) reporter gene fused to its promoter construct (Figure 6c–i). According to the observed pattern, AtECB2 was expressed very early, in germinating seeds (Figure 6c). In 3-day-old seedlings, GUS staining was primarily detected in the emerging cotyledons, as well as in the hypocotyls, with faint staining in the root (Figure 6d). In 14-day-old seedlings, GUS activity was detected in most of the plant, and was particularly strong in cotyledons (Figure 6f). GUS staining was also observed in the stem and silique, but not in the matured seeds (Figure 6e,h and i). Within the flower, GUS staining was intense in the sepals, the stamens and styles, but not in the petals (Figure 6g). GUS staining for transgenic lines showed that AtECB2 was widely expressed throughout the plant, and that higher expression levels were found in the cotyledon and in the seedling. Altogether, these findings show that AtECB2 was highly expressed in cotyledons and seedlings, which was in agreement with its putative roles in chloroplast biogenesis.

AtECB2 is required for the RNA editing of the accD site

Inspection of the DYW domain sequences from various plants (Salone et al., 2007) and early-diverging bryophytes (Ru¨ dinger et al., 2008) indicated that the domain was highly correlated with RNA editing. More recently, genetic evidence indicating that the DYW subgroup and RNA editing events are correlated have emerged (Zhou et al., 2008; Okuda et al., 2009; and Zehrmann et al., 2009). Therefore, we systematically checked the 34 editing sites that have been published (Tillich et al., 2005; Chateigner-Boutin and Small, 2007) in atecb2 mutants, by the direct sequencing of the RT-PCR products containing these sites. Of these published RNA editing sites, 33 sites were edited correctly in atecb2 mutants to be the same as those in the wild type grown under the same growth conditions. One editing site event, accD, did not occur, as shown in Figure 7a. To confirm the results, the PCR fragment containing the editing site from the wild-type or atecb2 cDNA was subjected to digestion of the restricted enzyme, EcoRI. The results showed that most of the RNA molecules were edited, whereas the edited molecules were hardly detected in atecb2 mutants (Figure 7b). We also evaluated the editing rates of the accD site in the wild type, complementary lines and atecb2 mutants. Our results showed that the rates in the wild type, and in the complementary lines, were 89.5% and 86.2%, respectively, under our culture conditions. In contrast, no molecules were edited in atecb2 mutants (Figure 7c), which suggested that the editing event was abolished in the mutant. Editing of codon 265 in wild-type accD transcripts encoding the β-subunit of carboxyltransferase (CT), the component of acetyl-CoA carboxylase (ACCase), resulted in a Leu→Ser residue change that was not detected in the mutant. The site was edited in the wild-type and the complemented plant under the same conditions (Figure 7). Therefore, we concluded that AtECB2 regulated the RNA editing of the site. Alignment analysis for the protein among the various species showed that leucine or similar amino acids were also conserved at the same position in the organisms (Sasaki et al., 2000). Thus, the non-editing of the site in atecb2 mutants might affect the functions of the protein.

Figure 7.

 RNA editing analysis for the chloroplast transcript accD in the wild type (WT), the complementary lines (TC) and atecb2 mutants.
(a) Sequence analysis for the accD RNA molecule from wild type, the complementary lines and from atecb2.
(b) dCAPS analysis for the editing site in the wild type, the complementary lines and in atecb2. Fragments originating from the wild type, the complementary lines and from atecb2 were amplified with the primers, and RT-PCR products were digested with the restriction enzyme EcoRI. The amplified sequence is shown, and the bold capital letter C displays the editing site. The underlined sequences refer to the primers.
(c) The RNA editing rate for the site accD in wild type, the complementary lines and in the atecb2 mutants.

Effects of the atecb2 mutation on the expression of chloroplast-encoded genes and the accumulation of the photosynthetic proteins

The expression of chloroplast-encoded genes is tightly linked with chloroplast developmental status, and their expression is carried out by two RNA polymerases of different origins: plastid-encoded polymerase (PEP) and nuclear-encoded polymerase (NEP). To investigate the effects of atecb2 mutation on these transcription processes, we compared the transcript profiles of various chloroplast-encoded genes between 10-day-old wild-type plants and atecb2 mutants (Figure 8). The five genes PsaB, PsbA, PsbB, PetD and RbcL, were selected as PEP-dependent genes (class I). AccD, Ycf2 and RpoA were chosen as NEP-dependent genes (class III). The four genes Rrn16, Rrn23, AtpB and AtpE were selected as both PEP- and NEP-dependent genes (class II) (Chateigner-Boutin et al., 2008; Myouga et al., 2008). Our results showed that the transcripts of class-I genes displayed 10–60% of the levels of mRNA produced by the wild-type transcripts. In contrast, the expression levels of class-III genes were higher in the mutant than in the wild type. Among the class-II genes, the expression levels of the two genes, Rrn16 and Rrn23, were about 20.72 and 16.79%, respectively, whereas the transcripts of the remaining two genes were about two and four times as high as those in the wild type. These results indicate that the expression of the plastid genes in Arabidopsis is affected by the atecb2 mutation.

Figure 8.

 Expression profiles of chloroplast-encoded genes in atecb2 mutants.
(a) Semiquantitative RT-PCR analysis of chloroplast genes between the wild type (WT) and atecb2 mutant.
(b) The expression profiles of plastid-encoded polymerase (PEP)-dependent chloroplast genes, between the WT and the atecb2 mutant; the expression profiles of nuclear-encoded polymerase (NEP)-dependent chloroplast genes, between the WT and the atecb2 mutant; and the expression profiles of both PEP- and NEP-dependent chloroplast genes, between the WT and the atecb2 mutant. The vertical axis shows the relative expressional level.

As the mutation affects chloroplast ultrastructure, we examined the accumulation of photosynthetic proteins in the mutant. The levels of the following photosynthetic proteins were examined by immunoblot assay in both the wild type and the atecb2 mutants (Figure 9): the large subunit of soluble rubisco proteins, RbcL (Krebbers et al., 1988); photosystem-I subunits, PsaD (Ihnatowicz et al., 2004) and PsaF (Haldrup et al., 2000); photosystem-II subunits, PsbA (Liere et al., 1995) and OEC33 (a product of the PsbO gene) (Jain et al., 1998); the cytb6f subunit, PetC (de Vitry et al., 1999); and the AtpB subunit (Inatomi, 1986). The photosynthetic proteins, including PsaD, PsaF, PsbA, RbcL and PetC were not detected in atecb2 mutants. The level of the AtpB subunit was reduced in atecb2 mutants compared with that of the wild type. Meanwhile, the level of OEC33, an extrinsic protein, was similar to that of the wild type. This indicated that the accumulation of most photosynthetic proteins was affected in atecb2 mutants, which is probably the consequence of defective chloroplast biogenesis.

Figure 9.

 Immunoblot analysis of photosynthetic proteins from the wild type (WT) and the atecb2 mutant. Total protein samples were prepared from 10-day-old seedlings, and were then separated by SDS-PAGE. Western blot analysis was carried out with the antisera. A 40-μg portion of total protein was loaded per lane.


In this study, we characterized the AtECB2 gene, which plays an essential role in proper chloroplast biogenesis. The lethal phenotype caused by the knock-out of AtECB2 demonstrates its essential role for chloroplast biogenesis. AtECB2 encodes a chloroplast-localized PPR protein. An investigation of all published RNA editing sites indicated that one site was abolished in atecb2 mutants, which suggests that AtECB2 participates in the RNA editing of this site.

AtECB2 is an essential gene for chloroplast biogenesis at an early stage

Molecular analysis of atecb2 plants showed that the gene was tagged with a T-DNA insertion, and that no transcripts of AtECB2 were present in atecb2 mutants (Figure 3c). Importantly, a 4876-bp wild-type genomic sequence, containing the AtECB2 gene as well as its 1936-bp upstream and 339-bp downstream sequences, was able to complement the atecb2 mutation, confirming that the AtECB2 gene was responsible for the albino phenotype. The successful complementary experiments also demonstrated that the 1936-bp upstream regions correctly drove the expression of the AtECB2 gene. Therefore, we constructed the regions with a fused GUS reporter gene to elucidate its organ-specific expression pattern. Histochemical assays displayed strong GUS staining in germinating seeds, especially in the cotyledons and seedlings (Figure 6), wherein the process of chloroplast differentiation was very active. The expression pattern was in agreement with its putative roles in chloroplast biogenesis at an early stage.

We characterized the phenotype of atecb2 mutants in detail to analyze the roles of AtECB2 in chloroplast biogenesis. The chloroplasts in atecb2 mutants were smaller than those in the wild type, and they contained only a few short appressed internal membranes. Thus, the complete loss of photosynthetic activity is probably the result of impaired chloroplast biogenesis in atecb2 mutants. Therefore, we conclude that AtECB2 is an essential gene for chloroplast biogenesis at an early stage. In addition, we investigated the accumulation of photosynthetic proteins in atecb2 mutants, as the mutation affects chloroplast ultrastructure. Like most albino mutants, the accumulation of thylakoid proteins was decreased substantially or absent (Figure 9). However, the defect in thylakoid protein accumulation in the mutant was not the result of the general impaired translation or protein translocation, as the accumulation of the extrinsic protein OEC33 was similar to that of the wild type. The large decrease in these proteins may result from the slight decrease of their transcripts (Figure S2); on the other hand, these thylakoid proteins may be incorrectly targeted, and are degraded because of the absence of a thylakoid membrane in the mutant.

The mutation of AtECB2 affects the coordinated gene expression of NEP and PEP initiation

Chloroplasts in higher plants possess at least two types of RNA polymerase: PEP and NEP (Kanamaru and Tanaka, 2004). It is generally accepted that NEP preferentially transcribes housekeeping genes, whereas PEP is predominantly involved in the transcription of photosynthetic genes (Allison et al., 1996; Hajdukiewicz et al., 1997). Previous reports have shown that mutations in these enzymes or their components result in delayed or completely inhibited chloroplast biogenesis (Hricováet al., 2006; Pfalz et al., 2006; Courtois et al., 2007; Garcia et al., 2008; Swiatecka-Hagenbruch et al., 2008). Therefore, chloroplast biogenesis requires the coordinated gene expression of NEP and PEP initiation. Recent reports have shown that the DG1 gene, encoding a PPR protein, was involved in the regulation of PEP transcription machinery, and affected the stability of PEP-dependent transcripts or the transcript assembly during the early stages of chloroplast biogenesis (Chi et al., 2008). Also, the phenotypes of reported Arabidopsis mutants, including ptac2/6/12 (Pfalz et al., 2006), atmurE (Garcia et al., 2008) and clb19 (Chateigner-Boutin et al., 2008), which are disrupted in those genes encoding the components of the PEP complex, are similar to that of the atecb2 mutant. The expression patterns of plastid-encoded genes in these mutants show defects in PEP-dependent transcription: i.e. the expression of class-I genes is downregulated, and the expression of class-III genes is upregulated. In atecb2 mutants, these plastid-encoded genes showed a similar expression pattern. These changes are probably the indirect consequences of abnormal chloroplast biogenesis. The expression of class-I genes was downregulated, which suggested that the levels of their transcripts were decreased or their stability was affected in atecb2 mutants. In addition, the transcripts of the class-III genes showed an over-accumulation in the mutant compared with those in the wild type. The data indicated that the NEP transcription machinery was functional in the mutant. The high activity of NEP observed in the mutant might be mediated by tRNAPGlu, which coordinates the activities of PEP and NEP through a feedback mechanism (Hanaoka et al., 2005). However, the transcriptional levels of class-II genes were found to be slightly upregulated or downregulated, which provided further evidence for the complexity underlying chloroplast transcription, as Chateigner-Boutin et al. (2008) suggested.

AtECB2 is required for chloroplast transcript AccD RNA editing

Comparative quantification of editing events has shown that the site is edited to a high degree (80–98%) in different tissues or under different growth conditions (Chateigner-Boutin and Small, 2007). Additionally, our data indicated that about 89.5% of molecules were edited in the wild type under our culture conditions (Figure 7c). The variance of the editing rate for this site in the wild type (Col.) may arise from the different growth conditions, and the high editing frequency for the site may be required for plant normal growth and development. CT is a component of ACCase, which is a key enzyme in fatty-acid synthesis. AccD encodes the β-polypeptide of CT. The editing event at codon 265 results in a change of amino acids in the AccD protein. Multiple alignments of the amino acid sequences deduced from the AccD gene also suggested that the leucine at this position was conserved among various plants and bacteria. Interruption of the AccD gene in Escherichia coli leads to the lethality of strains (Sasaki and Nagano, 2004). Therefore, AccD is an essential gene for E. coli. In plants that do not have a leucine codon at position 265, editing was shown to take place so as to create the leucine codon. The requirement of a leucine codon at a specific position suggests that AccD editing is necessary for several plants (Sasaki and Nagano, 2004). Biochemical experiments in vivo compared the unedited and edited recombinant enzymes, and their data indicated that the activity was found in the edited CT, but not in the unedited CT. Therefore, the editing event is needed for in vivo CT activity, and thereby for ACCase (Sasaki et al., 2000). So, the unedited AccD enzyme in atecb2 is probably inactive, which affects the consequent chloroplast biogenesis, although the transcripts of AccD still over-accumulated in the mutant. It is possible that a dysfunctional acetyl-CoA carboxylase in plastids contributes to defects in the early chloroplast biogenesis of the atecb2 mutant, which subsequently leads to an albino phenotype. Previous research in tobacco showed that the phenotype of leaves from transgenic tobacco carrying inactive AccD was variegated, and vesicle-like structures were found within the stroma of abnormal chloroplasts from transgenic leaves. This suggests the importance of plastid ACCase for maintaining leaf development and plastid structure (Kode et al., 2005). In atecb2 mutants, the exogenous expression of functional proteins translated from the edited AccD cDNA will provide direct data for this hypothesis.

Six PPR proteins have been reported to be involved in chloroplast transcript RNA editing in Arabidopsis, i.e. CRR4 (Kotera et al., 2005; Okuda et al., 2006), CRR21 (Okuda et al., 2007), CLB19 (Chateigner-Boutin et al., 2008), YS1 (Zhou et al., 2008), CRR22 and CRR28 (Okuda et al., 2009). Zehrmann et al. (2009) also isolated a PPR family member, MEF1, which is a mitochondria transcripts RNA-specific editing factor. Here, we identified another PPR protein, AtECB2, which is required for chloroplast transcript accD RNA editing and early chloroplast biogenesis. Among the RNA editing factors, three (CRR4, CRR21 and CLB19) belong to the E/E+ subgroup of the PPR family, and the others belong to the DYW subgroup. It is generally accepted that PPR motifs play a role in RNA recognition, and are bound with targeted RNA (Shikanai, 2006; Delannoy et al., 2007; Schmitz-Linneweber and Small, 2008). In Arabidopsis, all isolated PPR proteins that are RNA editing factors contain the highly conserved E/E+ domain. Therefore, the E/E+ domain is the common feature of the PPR proteins involved in RNA editing in Arabidopsis. This also suggests that this domain plays an important role in RNA editing, as Okuda et al. (2007, 2009) confirmed. The E/E+ domain without the adjacent DYW region was proposed to interact with other proteins, including the editing enzyme (Kotera et al., 2005; Okuda et al., 2007). However, the partners that the E/E+ domain interacts with have not yet been identified, nor has the function of the domain been confirmed by any direct biochemical evidence.

The DYW domain was proposed to be correlated with the presence of RNA editing, and contains a conserved HxExnCxxC motif that is similar to the active site of cytidine deaminases. This suggests that the DYW proteins have cytidine deaminase activity (Salone et al., 2007). However, no in vitro deaminases activity has been detected in isolated DYW proteins, to date. Okuda et al. (2009) discovered that the DYW motifs of CRR22 and CRR28 are dispensable for editing activity in vivo. They suggested that a DYW-containing protein in RNA editing may form heterodimers to perform cytidine deaminase activity by interacting with another DYW-containing protein (Okuda et al., 2009). Although no direct biochemical evidence was present to elucidate the DYW roles in the RNA editing process, genetic evidence showed that several genes, except for CRR2 (Hashimoto et al., 2003), are strongly in favor of a link between DYW proteins and editing events (Zhou et al., 2008; Okuda et al., 2009; Zehrmann et al., 2009). Our results from the present study demonstrate that another DYW-containing protein, AtECB2, is an editing factor for the accD site, which provides new genetic evidence supporting the link. The identification of an interacting partner of AtECB2 should be helpful to elucidate its functions in plastid RNA editing, as well as in early chloroplast biogenesis.

Experimental procedures

Plant material and growth conditions

Wild-type A. thaliana (ecotype Columbia) and mutants were grown under long-day conditions (16-h light/8-h dark), with a photon flux density of 120 μmol m−2 s−1 at a constant temperature of 22°C. For growth on agar plates, seeds were surface-sterilized and sown on a PNS medium containing 2% sucrose and 0.7% (w/v) phytoagar.

Mutant isolation and identification

To identify novel genes essential for chloroplast biogenesis at an early stage, genes highly or specifically expressed in early-stage seedlings were selected based on the gene expression information from the Genevestigator website (Zimmermann et al., 2004; A total of 34 genes, the products of which were thought to be chloroplast targeted, were identified (data not shown). Then phenotype screening was performed for these gene T-DNA insertional lines, which were obtained from the Arabidopsis Biological Resource Center (ABRC; The atecb2 mutant (SALK_112251) is one of the lines.

To determine the positions of T-DNA insertions for atecb2 mutants, genomic DNA for PCR was prepared as described by Sun et al. (2002). The flanking regions of the T-DNA insertions was amplified by PCR and sequenced with these primers: T-DNA flanking sequence, Lba1, 5′-TGGTTCACGTAGTGGGCCATCG-3′, and left primer, LP, 5′-CTCGTCTGGCTTCACAGAATC-3′; right primer, RP, 5′-GGAATTTTGGAATTCGGAGAC-3′. The wild-type line produced a PCR product of about 1000 bp (from LP to RP). The atecb2 mutant produced a band of about 600 bp (from RP to Lba1). Heterozygous lines produce both of the two bands.

Microscope observation

Transmission electron micrographs were obtained exactly as described by Motohashi et al. (2001). Small segments of the first primary leaf were from 14-day-old plants grown on a PNS medium supplement with 2% sucrose under normal conditions. The specimens were examined with a Hitachi H7500 transmission electron microscope (Hitachi,

Pigment detection and chlorophyll fluorescence analysis

Total chlorophyll was determined according to the method described by Lichtenthaler and Wellburn (1983). Extracts were obtained from 50 mg of fresh tissue from 14-day-old Arabidopsis seedlings, and were then homogenized in 100 mL of 100% acetone. Spectrophotometric quantification was carried out in a 721 spectrophotometer (Shanghai Third Analysis Instrument Factory). Chlorophyll-fluorescence measurements were performed using a pulse amplitude-modulated fluorometer (PAM 101; Walz, equipped with a data acquisition system to record fast changes (Meurer et al., 1996).

Mutant complementation

The 4876-bp wild-type AT1G15510 genomic fragment was amplified using LA-Taq polymerase (Takara, and the gene-specific primers: PPRGF, 5′-CGGGATCCTTCAGTTCTCTTCTGGGGATTTGCCAACCAG-3′, and PPRGR, 5′-GG CTGCAGGCTCTTTGCCGTTATCTTTGCTATCGGAATC-3′. After verification by sequencing, the fragment was cloned into a pCAMBIA1300 binary vector (CAMBIA; The transgenic lines for hygromycin resistance were further analyzed for their genomic background. Another primer set (HBLP, 5′-AGTATGCCTGTGTGGTGGGAATG-3′; HB3, 5′-CAGGAGTAATATTTTGATAGTG-3′) was used to identify the homozygous background for the T-DNA insertion site of AtECB2. The HB3 primer was located 300-bp downstream of the complementary fragment. No bands were amplified in the transgenic lines that had a homozygous background for the T-DNA insertion site of AtECB2, when the primer set was used. Three independent PCR reactions were repeated to confirm the homozygous background. At the same time, another PCR reaction with PPRGF and PPRGR primers was performed to monitor the quality of the DNA templates.

AtECB2 protein subcellular localization

To make the AtECB2TP:GFP construct, the transpeptide region was obtained by PCR amplification using two primers, AtECB2TF (5′-GGTACC ATGGCGTCTTCTGCTCAAAG-3′) and AtECB2TR (5′-CCATGG TAAGAACAGATAGTCCTTG-3′), cloned into the multiple clone sites of the vector pEGFP. The AtECB2TP:GFP fragment was cloned into a pMON530 binary vector. The 35S::GFP construct was used as a control. Both recombinant inserts were sequenced to ensure the authenticity of the transgenic fragment. The observation of GFP signals was performed as described by Motohashi et al. (2001).

AtECB2 promoter construction and histochemical analysis of GUS

The PAtECB2::GUS was made by amplifying the 1936-bp sequence upstream of the AtECB2 translation start sites using the primers AtECB2PF (5′-CGGAGCTCTTCAGTTCTCTTCTGGGGATTTGCCAACCAG-3′) and AtECB2PR (5′-CGCCATGG CGTAGGAGAGACAGAGCAAAGAACATGGTCACTCTC-3′), and subcloning the fragment into a pCAMBIA1301 binary vector. The different tissues of the transgenic lines were harvested into ice-cold 90% (v/v) acetone, incubated for 15–20 min on ice to permeabilize the tissue and then washed with a GUS assay, as described by Caissard et al. (1994). Tissues were examined using an Olympus SZ-CTV dissecting microscope interfaced with an Olympus DP70 digital camera ( and ACT-1 image-capture software.

Arabidopsis transformation and transformants selection

All binary vectors were introduced into Arabidopsis mediated by A. tumefaciens (Clough and Bent, 1998). Transformants were selected using 80 mg L−1 hygromycin B or 50 mg L−1 kanamycin.

RNA isolation, cDNA synthesis, RACE and RT-PCR

Total RNA from different tissues was isolated using the TRIZOL Reagent (Invitrogen, and DNase I treated by an RNeasy kit (Qiagen,, following the manufacturer’s instructions. The first-strand cDNA was synthesized with the revert-Aid first-strand cDNA synthesis kit (TOYOBO,, following the manufacturer’s instructions. The absence of DNA was confirmed by PCR prior to reverse transcription with random hexanucleotide primers. The full-length cDNA of the AtECB2 gene was obtained by using a SMART™ RACE cDNA amplification kit (Clontech,, following the supplier’s protocol. The two gene-specific primers, GSP1 (5′-CAGGCTTCACACCACCAACCCACAAC-3′) and GSP11 (5′-CGCGCTCTTCACATCACCACACTTC-3′) were used for 5′RACE, and the other two gene specific primers, GSP2 (5′-GGGGTCAAGGGTCGATGGTGGTAG-3′) and GSP22 (5′-GTGCGTGGTTGATTTGCTAGGCCGT-3′) were used for 3′RACE. To investigate the AtECB2 expression pattern, total RNA was isolated from roots, stems, leaves, inflorescences and 14-day-old seedlings, and semiquantitative RT-PCRs were performed using the specific primers as follows: AtECB2F (5′-ATGGCGTCTTCTGCTCAAAG-3′) and AtECB2R (5′-TAAGAACAGATAGTCCTTG-3′). The β-tubulin gene was used as a control, and the primer set was as follows: TubulinF (5′-GGACACTACACTGAAGGTGCTGAG-3′) and TubulinR (5′-CAAGCTGATGAACAGAGAGAGTTG-3′). For the expression analysis of AtECB2 in mutants, total RNA was prepared from 14-day-old mutants and wild types, and the two primers (AtECB2F and AtECB2R) were used. To analyze the transcripts of plastid genes, semiquantitative RT-PCR reactions were performed using the primers listed in Table S2. Reactions comprised 28 and 35 cycles of 1 min at 95°C, 1 min at 54°C, and 1 min at 72°C.

Quantitative real-time PCR analysis

Total RNA was extracted from 10-day-old wild-type and atecb2 plants, which were grown under a 16-h light/8-h dark regime, with a photon flux density of 120 mmol m−2 s−1, at 22°C. Quantitative real-time PCR amplifications were carried out in an ABI 7300 Real-Time PCR System (Applied Biosystems,, and the relative quantification of gene expression data was analyzed as described in Hricováet al. (2006). Primer pairs are listed in Table S2. The data set was normalized using β-tubulin as a control.

RNA editing analyses

Total RNA was extracted from 14-day-old plants. A total of 34 published RNA editing sites were amplified and sequenced by specific primers in Tillich et al. (2005) and Chateigner-Boutin and Small (2007). Each PCR fragment was sequenced twice by the two specific primers. The non-editing RNA sites in the mutant were sequenced another two times using independent plant samples. The RT-PCR products containing the accD editing site with specific primers, AccDF (5′-TTCAGAAAGAATCGAGAATT-3′) and AccDR (5′-AACAGCGTCAGTCAATCCTGTAG-3′), were digested with EcoRI, and analyzed on a 2.5% agarose gel. For analysis of the editing rate of the site accD, the sequence containing the editing sites from wild-type and atecb2 plants was amplified by PCR using the primers AccD-FW (5′-TCTTCTAGTTATTCCAATAATGTTGATCTTT-3′) and AccD-RV (5′-AGAGGTAAACATTGATTGGTAGCAT-3′). The RT-PCR products were ligated into pMD18-T vectors (Takara) and transformed into E. coli. PCR products from more than 100 independent clones, using the primers AccDF and AccDR, were digested with EcoRI and analyzed on a 2.5% agarose gel.

Immunological detection of photosynthetic proteins

Total proteins for immunological detection were extracted from 10-day-old wild-type and atecb2 mutants, as described previously (Motohashi et al., 2001). Proteins were separated by 12% or 15% SDS-PAGE, transferred electrophoretically to nitrocellulose filter membranes and were then immunoblotted with various antibodies. Antibodies were detected using an enhanced chemiluminescence (ECL; General Electric Company, following the manufacturer’s instructions. All polyclonal antibodies in this study, except for RbcL, were obtained from Agrisera ( The RbcL antibody was presented by Dr Mi Hua-Ling from the National Key Laboratory of Plant Molecular Genetics, Institute of Plant Physiology and Ecology, at the Chinese Academy of Sciences.


We would like to thank ABRC Bioresources, who kindly offered the transgenic Arabidopsis lines (SALK_112251). We thank Mrs Hui-Qi Zhang from SHNU for her skillful technical assistance in transmission electron microscopy. We are grateful to Dr Hua-Ling Mi and Ms Meng-Meng Kong from SIPPE for help in immunology experiments. This work was supported by grants from the National Basic Research Program of China (2009CB118504), National Science Foundation of China (30530100) and by a Leading Academic Discipline Project of Shanghai Municipal Education Commission (J50401).