The rice pentatricopeptide repeat protein RF5 restores fertility in Hong-Lian cytoplasmic male-sterile lines via a complex with the glycine-rich protein GRP162

Abstract

The cytoplasmic male sterility (CMS) phenotype in plants can be reversed by the action of nuclear-encoded fertility restorer (Rf) genes. The molecular mechanism involved in Rf gene-mediated processing of CMS-associated transcripts is unclear, as are the identities of other proteins that may be involved in the CMS-Rf interaction. In this study, we cloned the restorer gene Rf5 for Hong-Lian CMS in rice and studied its fertility restoration mechanism with respect to the processing of the CMS-associated transcript atp6-orfH79. RF5, a pentatricopeptide repeat (PPR) protein, was unable to bind to this CMS-associated transcript; however, a partner protein of RF5 (GRP162, a Gly-rich protein encoding 162 amino acids) was identified to bind to atp6-orfH79. GRP162 was found to physically interact with RF5 and to bind to atp6-orfH79 via an RNA recognition motif. Furthermore, we found that RF5 and GRP162 are both components of a restoration of fertility complex (RFC) that is 400 to 500 kD in size and can cleave CMS-associated transcripts in vitro. Evidence that a PPR protein interacts directly with a Gly-rich protein to form a subunit of the RFC provides a new perspective on the molecular mechanisms underlying fertility restoration.

Figure 1.

Map-Based Cloning of Rf5. (A) The Rf5 locus was initially mapped to chromosome 10 between the SSR markers RM1108 and RM5373 and subsequently narrowed down to a 67-kb interval between the markers RM6469 and RM25659. Subcloning and sequencing of Milyang23 BAC clone 68F6. (B) Analysis of the percentage of fertile pollen from T1 progeny of the rf5 line YTA containing the transgenic Rf5 complementation construct (χ² = 0.013 for 1:1, P < 0.05). (C) Pollen fertility was assessed by 1% I₂-KI staining. Darkly stained pollen is fertile, and lightly stained pollen is sterile. (D) RNA gel blot analysis of the CMS rf5 line YTA, the Rf5 transgenic YTA lines T₁-6, and the Rf5 line NILF₁ using orfH79 as a probe and an actin probe as a control; an ethidium bromide stain of the gel is shown to confirm equal RNA loading. The RNA was extracted from seedling leaves. (E) HptII gene in the transgenic vector was amplified and the PCR products were loaded on a 1.5% agarose gel to show the cosegregation of the HptII site and fertile plants among F1 progeny. M, marker DL2000, NC, negative control.

Figure 2.

Interaction between RF5 and CMS-Associated Transcripts. (A) EMSA with MBP-RF5 and MBP tag using the atp6-orfH79 RNA probe. (B) The atp6-orfH79 intragenic region from 978 to 1285 nucleotides, including parts of the atp6 and orfH79 sequences, was in vitro transcribed as a CMS RNA probe for EMSA. The intragenic region including parts of the atp6 and orfH79 sequences was divided into three parts, A, B, and C, for a yeast three-hybrid assay.

Figure 3.

Interaction between GRP162 and RF5. (A) GRP162 and RF5 were fused to AD and BD vectors, respectively, in a yeast two-hybrid assay to show the interaction between GRP162 and RF5. (B) His-tagged GRP162 was conjugated to a HisTrap column, and recombinant MBP-RF5 was applied to the column. The eluted fractions 7 to 13 were loaded into lanes 4 to 10 of SDS-PAGE and visualized with Coomassie blue staining. M, prestained marker 0671 (Fermentas). (C) Anti-His and anti-MBP immunoblots of elution fraction 9 from the pull-down assay and controls. (D) BiFC assay for detecting molecular interactions between RF5-YFP(N) and GRP162-YFP(C). (E) BiFC assay for detecting interactions between GRP162 molecules. GRP162-YFP(N) and GRP162-YFP(C) were coexpressed in onion epidermal cells.

Figure 4.

The Expression of GRP162. Immunodetection of GRP162 accumulation in mitochondria with an anti-GRP162 antibody. An antibody against RPL12, a ribosomal protein cell that is associated with mitochondria, was used as a control. An antitubulin antibody was used as a control to rule out contamination. Twenty micrograms of prepared mitochondria (Mit.) cell and 30 μg of total cell protein (Tot.) from YTA and NILF₁ were loaded.

Figure 5.

Interaction between the GRP162 and CMS-Associated Transcripts. (A) EMSA of His-GRP162 and His-tag using a fixed concentration (0.5 nM) of the CMS RNA probe (diagrammed in Figure 2B) with a range of protein concentrations and His-tag as a control. (B) EMSA of the His-tagged RRM domain of Rf5 (His-RRM) using a fixed concentration (0.5 nM) of the RNA probe (diagrammed in Figure 2B) with a range of protein concentrations. (C) EMSAs with His-GRP162 and His-tag were performed using a fixed concentration (0.3 nM) of fragment A, B, or C as the RNA probe (diagrammed in Figure 2B) with a range of protein concentrations. (D) β-Galactosidase activity for the three-hybrid assay and the details of rows 1 to 8. MS2-A⁺, MS2-B⁺, and MS2-C⁺ (indicated by A, B, and C, respectively) were ligated into pIIIMS2-2 to produce sense strand RNA probes. MS2-A⁻MS2-B⁻and MS2-C⁻ denote reverse ligations that produce antisense strand RNA probes. (E) After the transformants were selected for the presence of the plasmids, colonies were assayed for HIS3 reporter activity and 3-AT competition.

Figure 6.

Identification of the RFC in Mitochondria. (A) RT-PCR for atp6 and orfH79 to detect RNA-protein immunoprecipitates. Actin and 26S RNA were used as controls. cDNA was used as template in the positive control (PC), and RNA was used as template in the negative control (NC). M, marker DL2000. (B) Coimmunoprecipitation assay with anti-GRP162 immunoprecipitates detected with anti-RF5 antibody. (C) Coomassie blue staining of 100 μg of mitochondria separated by native polyacrylamide gel electrophoresis. Complex I (I), Complex V (V), and Complex III₂ (III₂) are visible in the gel. (D) Anti-RF5 and anti-GRP162 protein gel blots to detect the RF5 complex by BN-PAGE. (E) Anti-RF5 and anti-GRP162 protein gel blots to detect F1 hybrid mitochondrial fractions resulting from size-exclusion chromatography with Superdex 200. (F) Protein standards, including ferritin (440 kD, 880 kD as a dimer), catalase (220 kD), and BSA (68 kD), were run on Superdex 200. [See online article for color version of this figure.]

Figure 7.

In Vitro Cleavage Activity Assay for RF5 and GRP162. (A) Transcribed RNAs encompassing the intragenic region of atp6-orfH79 from 19 to 1285 nucleotides were incubated in vitro with recombinant RF5 and GRP162 to assess cleavage activity. (B) An RNA cleavage assay using RNA and protein from mitochondrial extract from the HL-CMS line together with recombinant RF5 and GRP162. Reverse transcription for time-dependent quantitative PCR analysis of the expression of atp6-orfH79. The vertical line (the y axis) corresponds to the relative transcript level, and each bar represents the mean ± sd (n = 3). (C) An RNA cleavage assay with total mitochondrial RNA as substrate for fractions 2 to 11 from the size-exclusion chromatography and with PBS as the control in fraction 0 (line 1). The RNA was analyzed by RNA gel blotting with orfH79 as a probe.

Figure 8.

Analysis of GRP RNAi Lines. (A) Quantitative RT-PCR analysis of GRP162 expression levels in Rf5 (NIL) and GRP-RNAi lines. The data are shown as mean ± sd (n = 3). (B) RNA gel blot analysis of the CMS-associated transcript atp6-orfH79 in the YTA, Rf5 (NIL), and GRP-RNAi lines. Ethidium bromide staining of the gel confirms equal RNA loading.

Figure 9.

The Processing Site of atp6-orfH79 and Secondary Structure Prediction. (A) The processing site of the atp6-orfH79 transcript was determined using a primer extension assay as indicated with the readable sequence. (B) The secondary structure of atp6-orfH79 was predicted using RNAdraw 1.1; the processing site (at 1169 nucleotides) is marked with an arrow.