Accumulation of the RNA polymerase subunit RpoB depends on RNA editing by OsPPR16 and affects chloroplast development during early leaf development in rice

Abstract

Plastid-encoded genes are coordinately transcribed by the nucleus-encoded RNA polymerase (NEP) and the plastid-encoded RNA polymerase (PEP). Resulting primary transcripts are frequently subject to RNA editing by cytidine-to-uridine conversions at specific sites. The physiological role of many editing events is largely unknown. Here, we have used the CRISPR/Cas9 technique in rice to knock out a member of the PLS-DYW subfamily of pentatricopeptide repeat (PPR) proteins. We found that OsPPR16 is responsible for a single editing event at position 545 in the chloroplast rpoB messenger RNA (mRNA), resulting in an amino acid change from serine to leucine in the β-subunit of the PEP. In striking contrast to loss-of-function mutations of the putative orthologue in Arabidopsis, which were reported to have no visible phenotype, knockout of OsPPR16 leads to impaired accumulation of RpoB, reduced expression of PEP-dependent genes, and a pale phenotype during early plant development. Thus, by editing the rpoB mRNA, OsPPR16 is required for faithful plastid transcription, which in turn is required for Chl synthesis and efficient chloroplast development. Our results provide new insights into the interconnection of the finely tuned regulatory mechanisms that operate at the transcriptional and post-transcriptional levels of plastid gene expression.

Keywords: Chl biosynthesis; RNA editing; RNA polymerase; chloroplast development; plastid; transcription.

Fig. 1

Generation and characterization of rice (Oryza sativa) osppr16 mutants. (a) Strategy I and (b) Strategy II for CRISPR/Cas9‐dependent gene editing of OsPPR16. Mutants obtained with Strategy I and Strategy II and selected for in‐depth characterization were named osppr16d and osppr16s, respectively. (c) Sequences around the two CRISPR targets (Target1 and Target2) in wild‐type (WT) and osppr16d. PAM, protospacer adjacent motif. (d) Sequence around Target3 in WT and osppr16s. (e) osppr16d, osppr16s, and WT plants at the three‐leaf stage. Bar, 10 cm. (f) osppr16d, osppr16s and WT plants at the five‐leaf stage. Bar, 10 cm. (g) osppr16d, osppr16s, and WT plants at seed set. Bar, 10 cm. (h, i) Chloroplast ultrastructure of (h) WT and (i) osppr16d at the three‐leaf stage. Bars, 1 μm. (j, k) Chloroplast ultrastructure of (j) WT and (k) osppr16d plants at the five‐leaf stage. Bars, 1 μm.

Fig. 2

Analysis of Chl synthesis in rice (Oryza sativa) wild‐type (WT) and osppr16 mutant plants. (a–c) Chl content of leaves at different development stages in osppr16 and WT plants. The contents of (a) Chla, (b) Chlb, and (c) the total Chl are shown. TLS, three‐leaf stage; FLS, five‐leaf stage; SRS, seed ripening stage. Error bars indicate SD (n = 4), *, P < 0.05; **, P < 0.01 (Student’s t‐test). (d) Time course of de‐etiolation of WT and osppr16 seedlings. Plants were photographed 0, 3, 6, 12, 24, and 48 h after light treatment. Bars, 10 cm. (e) Chl contents in leaves of WT and osppr16d during the de‐etiolation time course. Error bars indicate SD (n = 4), **, P < 0.01 (Student’s t‐test).

Fig. 3

Domain structure and subcellular localization analysis of OsPPR16. (a) Structure of OsPPR16. CTP, chloroplast transit peptide. (b) Subcellular location of OsPPR16 in rice protoplasts. YFP, yellow fluorescent protein. Bars, 5 µm. (c) Co‐localization of OsPPR16‐GFP (yellow) and PEND‐CFP (cyan) in chloroplast nucleoids. Note that the GFP fluorescence was detected in the YFP channel, and therefore is represented as yellow color. GFP, green fluorescent protein; CFP, cyan fluorescent protein. Bars, 5 µm.

Fig. 4

Editing analysis of rpoB‐545 in rice (Oryza sativa) wild‐type (WT), osppr16, RNA interference (RNAi), and complemented plants. (a) Editing efficiency of rpoB‐545 in WT and osppr16 leaves at three‐leaf, five‐leaf, and seed ripening stages. SRS, seed ripening stage. The arrows refer to the editing site. (b) Relative expression of OsPPR16 in RNAi transgenic plants, as determined by quantitative real‐time PCR. OsTPI was used as reference gene. (c) Editing efficiency of rpoB‐545 in OsPPR16 RNAi lines. EE, editing efficiency. (d) Assay for the Cas9 gene by PCR in osppr16d T₁ plants. M, marker. Vector, pYLCRISPR‐osppr16d. The osppr16d‐36‐6 plant was confirmed as Cas9‐free line, and subsequently used to conduct complementation analysis. (e) Analysis of Chl content, transfer DNA (T‐DNA) presence, and RNA editing efficiency in WT, osppr16d, and T₁ seedlings from a T₀ complemented line at the three‐leaf stage. (f) Phenotypes of WT, osppr16d, and complemented T₁ seedlings (P_OsPPR16::OsPPR16/osppr16). +, positive T₁ seedlings segregated from a T₀ complementation line; −, negative T₁ seedlings segregated from a T₀ complementation line. Bar, 10 cm.

Fig. 5

Protein and messenger RNA accumulation of plastid‐encoded RNA polymerase (PEP) core subunits in rice (Oryza sativa) wild‐type (WT) and osppr16 mutant plants. (a) Immunoblot analysis of RpoB in WT and osppr16 leaves at the three‐leaf and five‐leaf stages. An antibody against the heat‐shock protein HSP82 was used as an internal control. (b, c) Relative expression levels of the four genes encoding PEP core subunits in WT and osppr16 leaves at (b) the three‐leaf stage and (c) the five‐leaf stage, as determined by quantitative real‐time PCR. OsTPI was used as reference gene. Error bars indicate SD (n = 3); *, P < 0.05; **, P < 0.01 (Student’s t‐test).

Fig. 6

RNA editing at rpoB‐545 likely facilitates proper protein folding of the plastid‐encoded RNA polymerase (RNAP) β‐subunit. (a) Partial amino acid sequence alignment of RNAP β‐subunits of five bacterial species reveals a conserved leucine at the position corresponding to the edited codon (codon 182) of the rice chloroplast β‐subunit. By contrast, sequence alignment of chloroplast β‐subunits of four plant species reveals a serine at these positions (denoted by the asterisk). T. aquaticus, Thermus aquaticus; E. coli, Escherichia coli; M. tuberculosis, Mycobacterium tuberculosis; N. calcicola, Nostoc calcicola; G. violaceus, Gloeobacter violaceus; O. sativa, Oryza sativa; Z. mays, Zea mays; N. tabacum, Nicotiana tabacum; A. thaliana, Arabidopsis thaliana. (b) The crystal structure of T. aquaticus RNAP‐promoter open complex shows that L197 is located within the β‐lobe domain (dashed box) of the RNAP β‐subunit. The RNAP α, β, β′ and ω subunits are shown in cyan, gray, dark gray and blue, respectively, the promoter DNA is shown in red, and the σ‐factor is shown in yellow. (c) Amino acid L197 is buried inside the hydrophobic core of the β‐lobe domain. Numbers of amino acid residues correspond to the positions in the T. aquaticus RpoB protein. The equivalent positions in the O. sativa protein are given in parentheses.

Fig. 7

Transcript accumulation of plastid‐encoded RNA polymerase (PEP)‐dependent genes in rice (Oryza sativa) wild‐type (WT) and osppr16 mutant leaves. (a, b) Quantitative real‐time PCR analysis of the expression of PEP‐dependent genes in WT and osppr16 leaves at the (a) three‐leaf stage and (b) five‐leaf stage. OsTPI was used as reference gene. Error bars indicate SD (n = 3); *, P < 0.05; **, P < 0.01 (Student’s t‐test) in (b); all P < 0.01 in (a). (c) Immunoblot analysis of selected proteins encoded by PEP‐dependent genes in WT and osppr16 leaves at the three‐leaf stage. Samples of 25 μg total cellular protein extracted from osppr16d leaves were loaded and compared with a dilution series of WT protein extracts. The heat shock protein HSP82 was used as an internal control. (d) Immunoblot analysis of the same proteins at the five‐leaf stage. Samples of 25 μg total protein from WT and osppr16d leaves were analyzed by Western blot analysis. HSP82 was used as an internal control.

Fig. 8

Expression of Chl biosynthesis‐related genes in rice (Oryza sativa) wild‐type (WT) and osppr16 mutant plants. (a) Northern blot analysis of accumulation of tRNA^Glu in osppr16 mutants and WT plants at the three‐leaf stage and five‐leaf stage. 28S rRNA accumulation was used as internal loading control. (b, c) Quantitative real‐time PCR analysis of the expression of nuclear genes related to tetrapyrrole synthesis at the (b) three‐leaf stage and (c) five‐leaf stage. OsTPI was used as reference gene. Error bars indicate SD (n = 3); *, P < 0.05; **, P < 0.01 (Student’s t‐test).

Fig. 9

A working model for the role of OsPPR16 during chloroplast development in rice. During early chloroplast development, OsPPR16 is particularly important to edit rpoB transcripts to ensure maximum RpoB accumulation, thereby producing an active plastid‐encoded RNA polymerase (PEP) pool that is capable of efficiently transcribing PEP‐dependent genes (PDGs). The accumulation of tRNA^Glu and proteins encoded by other PEP‐dependent genes is required to meet the demand for Chl and protein synthesis during the rapid development of proplastids into chloroplasts. Loss of OsPPR16 leads to an extremely low level of tRNA^Glu (and proteins encoded by other PEP‐dependent genes, PPDGs), thereby limiting Chl biosynthesis and protein synthesis (and, potentially, also protein stability in the absence of sufficient pigments for incorporation into Chl‐binding proteins) in the developing chloroplasts, thus causing the pale phenotype during early leaf development. Lower demands for PEP activity at later developmental stages and gradually increasing PEP accumulation in the mutants alleviate the phenotype and facilitate leaf greening. See the Discussion section for more details.