A pentatricopeptide repeat protein restores nap cytoplasmic male sterility in Brassica napus

Abstract

Two forms of male-sterile cytoplasm, designated nap and pol, are found in the oilseed rape species, Brassica napus. The nap cytoplasm is observed in most B. napus varieties, and it confers male sterility on a limited number of cultivars that lack the corresponding restorer gene, Rfn. In the present study, using linkage analysis in combination with 5652 BC1 progeny derived from a cross between a nap cytoplasmic male sterility (CMS) line 181A and a restorer line H5, we delimited the Rfn gene to a 10.5 kb region on chromosome A09, which contained three putative ORFs. Complementation by transformation rescue revealed that the introduction of ORF2, which encodes a pentatricopeptide repeat (PPR) protein, resulted in the recovery of fertility of nap CMS plants. Expression analysis suggested that the Rfn was highly expressed in flower buds and it was preferentially expressed in the tapetum and meiocytes during anther development. Further RNA gel blots and immunodetection suggested that the Rfn gene may play a complicated role in restoring the nap CMS. Our work laid the foundation for dissecting the molecular basis of CMS fertility restoration and the nuclear-mitochondrial interactions in CMS/Rf systems.

Keywords: Brassica napus; cytoplasmic male sterility; map-based cloning; mitochondria; pentatricopeptide repeat protein; restorer of fertility (Rf).

Fig. 1.

Map-based cloning of the Rfn gene for nap CMS in Brassica napus. (A) Molecular mapping of the Rfn locus using a BC₁ population containing 5652 individuals. The numbers of recombinants between the markers and Rfn are shown. (B) Physical map based on a sequenced BAC. The Rfn locus was narrowed down to a 10.5 kb region. (C) The predicted genes from http://www.softberry.com. It contained two complete CDSs (ORF1 and ORF2) and a partial CDS (ORF3). (D) Morphology of flowers and pollen grains of a nap CMS plant and two hemizygous Rfn candidate transgenic plants. Dissected flowers were photographed with a digital camera. Pollen grains were stained with 1% aceto-carmine staining solution. Scale bar=100 μm. (E) Co-segregation analysis of the introduced DNA and fertile plants among the T₁-20 progeny. F, fertile plant; S, sterile plant; N, negative control; P, positive control; M, DL2000 molecular markers. (This figure is available in colour at JXB online.)

Fig. 2.

Combinations of amino acid residues at positions 5 and 35 of PPR motifs and the RNA sequence predicted to be recognized by these combinations for RFN and rfn proteins. ‘N’ represents that the recognized RNA base is unknown; ‘R’ represents that the recognized RNA base is A or G; ‘Y’ represents that the recognized RNA base is U or C. The same or different amino acid residue combinations in each PPR motif between RFN and rfn and their recognized RNA bases are indicated by different shading. (This figure is available in colour at JXB online.)

Fig. 3.

The expression patterns of Rfn (rfn) by qRT–PCR and RNA in situ hybridization. (A) qRT–PCR analysis of Rfn and rfn expression in the selective tissues of parent lines H5 and 181A. R, root; S, stem; L, leaves; FB, flower buds; OF, opening flowers; GS, green siliques. The expression levels were normalized to BnActin. The values are presented as the means ±SD (n=3 biological replicates). (B–G) The expression patterns of Rfn by RNA in situ hybridization in the anther of the fertile parent H5 at different developmental stages detected by the antisense probe (B–F) and sense probe (G). MC, meiotic cell; MMC, microspore mother cell; Msp, microspores; PG, pollen grains; T, tapetum; Tds, tetrads. Scale bars=50 μm. (This figure is available in colour at JXB online.)

Fig. 4.

Subcellular localization of the N-terminal 44 residue signal peptide of RFN fused with green fluorescent protein (GFP). (A) The protoplast showed a green fluorescent signal at 488 nm. (B) The same protoplast showed a red fluorescent signal (stained by MitoTracker Red CMX-Ros) at 580 nm. (C) The bright-field image. (D) The merged image of (A), (B), and (C). Scale bars=10 µm. (This figure is available in colour at JXB online.)

Fig. 5.

RNA gel blot analysis of the orf222/nad5c/orf139 transcripts with probes orf222 and orf139, respectively, and immunodetection of ORF222 from CMS, restorer line, and transgenic fertility-restored plants. (A) The structure of nap CMS-associated transcripts orf222/nad5c/orf139. The shaded portion indicated the orfB homologous region. (B) RNA gel blots analysis with probes orf222 (left panel) and orf139 (right panel), respectively. The estimated sizes of the different transcripts are shown in kilobases. Ethidium bromide stain of the gel confirms equal RNA loading. (C) Immunodetection of CMS protein expression in T₁ progeny with an anti-ORF222 antibody. Equal amounts of total protein were loaded in each lane stained by Coomassie Brilliant Blue. (This figure is available in colour at JXB online.)