Pentatricopeptide repeat protein MITOCHONDRIAL STABILITY FACTOR 3 ensures mitochondrial RNA stability and embryogenesis

Abstract

Gene expression in plant mitochondria is predominantly governed at the post-transcriptional level and relies mostly on nuclear-encoded proteins. However, the protein factors involved and the underlying molecular mechanisms are still not well understood. Here, we report on the function of the MITOCHONDRIAL STABILITY FACTOR 3 (MTSF3) protein, previously named EMBRYO DEFECTIVE 2794 (EMB2794), and show that it is essential for accumulation of the mitochondrial NADH dehydrogenase subunit 2 (nad2) transcript in Arabidopsis (Arabidopsis thaliana) but not for splicing of nad2 intron 2 as previously proposed. The MTSF3 gene encodes a pentatricopeptide repeat protein that localizes in the mitochondrion. An MTSF3 null mutation induces embryonic lethality, but viable mtsf3 mutant plants can be generated through partial complementation with the developmentally regulated ABSCISIC ACID INSENSITIVE3 promoter. Genetic analyses revealed growth retardation in rescued mtsf3 plants owing to the specific destabilization of mature nad2 mRNA and a nad2 precursor transcript bearing exons 3 to 5. Biochemical data demonstrate that MTSF3 protein specifically binds to the 3' terminus of nad2. Destabilization of nad2 mRNA induces a substantial decrease in complex I assembly and activity and overexpression of the alternative respiratory pathway. Our results support a role for MTSF3 protein in protecting two nad2 transcripts from degradation by mitochondrial exoribonucleases by binding to their 3' extremities.

Figure 1

A null allele of the Arabidopsis MTSF3 gene confers embryo lethality. A, Schematic diagram of the MTSF3 gene structure with the position of the SAIL_359_F11 T-DNA insertion. The length of the putative mitochondrial targeting sequence (mTP) was predicted using TargetP and is shown as a black box. B, Open siliques showing the seeds produced from wild-type (Col-0), heterozygous, and functionally-complemented mtsf3 plants. Arrows point toward white seeds that are observed in siliques from mtsf3 heterozygous plants. Bars = 50 μm. C, Confocal images of embryos dissected from immature seeds (6 weeks after sowing) showing the phenotype of mtsf3 mutant embryos compared to an embryo contained in a green seed. Bars = 10 μm or 50 μm.

Figure 2

Partially-complemented pABI3::MTSF3 plants display a globally retarded growth phenotype. Photograph of 5-week-old plants showing the reduced size of two partially-complemented (pABI3::MTSF3#1 and #2) mtsf3 mutant plants as compared with the wild type (Col-0) and a fully-complemented (35S::MTSF3::GFP) mtsf3 homozygous mutant.

Figure 3

Respiratory complex I is decreased in partially-complemented mtsf3 plants. A, BN-PAGE analysis of mitochondrial complex I accumulation in pABI3::MTSF3 plants. In-gel activity staining revealing the NADH dehydrogenase activity of complex I is presented in the left panel. Detection of complex I has also been performed on BN-PAGE blots with antibodies to the mitochondrial CA2 (carbonic anhydrase 2) subunit as shown on the right panel. Holocomplex I is indicated by an arrowhead. B, Analysis of mitochondrial proteins steady-state levels. Crude membrane extracts from the indicated genotypes were separated by SDS–PAGE and probed with antibodies to subunits of complex I (Nad7 and Nad9), complex III (RISP), complex IV (Cox2), the ATP synthase (ATPβ), cytochrome c (CYT C), and the alterative oxidase (AOX). Porin was used as protein loading control. Dilution series of proteins extracted from the wild type (Col-0) was used for signal comparison.

Figure 4

Mature nad2 mRNA strongly under-accumulate in partially-complemented mtsf3 mutants. The steady-state levels of mature mitochondrial mRNAs were measured by RT-qPCR in Col-0 and partially-complemented mtsf3 plants. The histograms show log₂ ratios of pABI3::MTSF3 to wild-type. A single RT-qPCR was considered for mRNAs carrying no introns, whereas the accumulation of individual exons was analyzed for intron-containing transcripts. Three biological replicates and three technical replicates were used per genotype; standard errors are indicated. The data were normalized to the nuclear 18S rRNA gene.

Figure 5

The stability of the nad2 exon 3–5 precursor is compromised in partially-complemented mtsf3 mutants. A, Schematic representation of nad2 exon 1–2 and nad2 exon 3–5 precursor transcripts and positions of the primers used for RT-qPCR analysis. B, RT-qPCR measuring the steady-state levels of nad2 exon 1–2 and nad2 exon 3–5 precursor RNAs in Col-0 and partially-complemented mtsf3 plants. The histograms show log₂ ratios of pABI3::MTSF3 plants to wild-type. RT-qPCR reactions were performed with the indicated primer pairs. Three biological replicates and three technical replicates were used per genotype; standard errors are indicated. C, Schematic representation of the two nad2 precursor transcripts, which are fused by one trans-splicing event to produce the mature (m) nad2 mRNA. Boxes indicate exons. Introns and 5′ and 3′ UTRs are shown as thick and thin lines, respectively. The probes used to interpret RNA gel blot results are also indicated. D, RNA gel blots showing the accumulation profiles of nad2 transcripts in the two partially-complemented mtsf3 mutants compared to WT (Col-0). Used probes are indicated below hybridization results. Ethidium bromide staining of ribosomal RNAs is shown below the blots and serves as a loading control. m: mature mRNAs.

Figure 6

The MTSF3 protein binds specifically to the 3′ region of the nad2 mRNA. A, Putative binding sites of MTSF3 were predicted based on the amino acid residues at positions 5 and 35 of PPR motifs 3–16. Repeats are listed from N to C-terminus. The obtained combinations were then used to calculate the probabilities of nucleotide recognition by each individual PPR repeat according to the PPR code. The sequence logo depicting these probabilities was obtained with http://weblogo.berkeley.edu/. B, Prediction of potential MTSF3 binding sites within the Arabidopsis Col-0 mitochondrial genome. The 10 most-probable MTSF3 binding sequences matching mitochondrial mRNA transcripts are shown. The arrowhead marks the site located in the 3′ region of nad2, shown in subsequent experiments to be the RNA target of MTSF3. The P-values were determined with the FIMO program. C, The MTSF3 protein specifically associates with the 3′ end extremity of the nad2 mRNA in vivo. Total extracts from transgenic Arabidopsis cells expressing an MTSF3-GFP fusion and control untransformed PSB-D cells were used for immunoprecipitation with an anti-GFP antibody. Coimmunoprecipitated RNAs were analyzed by RT-qPCR using primer pairs of the indicated mitochondrial transcript regions. Amplification products covering nad2 precursor transcripts are shown in the orange box above the histogram. nad2 exon 5 3′-UTR (PCR #15) shows the enrichment obtained with a primer pair targeting nad2 3′ extremity. Immunoblot results of total extracts (Input), flow-through (FT), and immunoprecipitated (IP) fractions using the GFP antibody is presented. D, SDS protein gel stained with Coomassie blue showing the purity of the HIS-MTSF3 fusion protein overexpressed and purified from Escherichia coli. M: protein size marker (kDa). E, Electrophoretic mobility shift assay showing the association of MTSF3 to the 3′ region of nad2 in vitro. EMSA assays were performed with purified HIS-MTSF3 and the RNA probes depicted in panel F. For each probe, the reactions contained (from left to right) 0, 200, 400, and 800 nM of pure HIS-MTSF3 protein, 100 pM of gel-purified RNA probe, and 2 mg/mL heparin. U: unbound probe; B: bound probe. F, Diagram of the nad2 mRNA with the relative positions of the RNA probes (numbered from 1 to 4) used in EMSA assays. The MTSF3 binding site is indicated as MTSF3 BS.