TFIIB-Related Protein BRP5/PTF2 Is Required for Both Male and Female Gametogenesis and for Grain Formation in Rice

Abstract

Transcription factor IIB (TFIIB) is a general transcription factor for RNA polymerase II, exerting its influence across various biological contexts. In the majority of eukaryotes, TFIIB typically has two homologs, serving as general transcription factors for RNA polymerase I and III. In plants, however, the TFIIB-related protein family has expanded greatly, with 14 and 9 members in Arabidopsis and rice, respectively. BRP5/pollen-expressed transcription factor 2 (PTF2) proteins belong to a subfamily of TFIIB-related proteins found only in plants and algae. The prior analysis of an Arabidopsis atbrp5 mutant, characterized by a T-DNA insertion at the 5' untranslated region, demonstrated the essential role of BRP5/PTF2 during the process of pollen germination and embryogenesis in Arabidopsis. Using a rice transformation system based on CRISPR/Cas9 technology, we have generated transgenic rice plants containing loss-of-function frameshift mutations in the BRP5/PTF2 gene. Unlike in the Arabidopsis atbrp5 mutant, the brp5/ptf2 frameshift mutations were not transmitted to progeny in rice, indicating an essential role of BRP5/PTF2 in both male and female gamete development or viability. The silencing of rice BRP5/PTF2 expression through RNA interference (RNAi) had little effect on vegetative growth and panicle formation but strongly affected pollen development and grain formation. Genetic analysis revealed that strong RNAi silencing of rice BRP5/PTF2 was still transmissible to progeny almost exclusively through female gametes, as found in the Arabidopsis atbrp5 knockdown mutant. Thus, reduced rice BRP5/PTF2 expression impacted pollen preferentially by interfering with male gamete development or viability. Drawing upon these findings, we posit that BRP5/PTF2 assumes a distinct and imperative function in the realm of plant sexual reproduction.

Keywords: TFIIB; gametogenesis; pollen development; rice; transcription factor.

Figure 1

Rice BRP5 gene structure and protein sequence. (A) The OsBRP5 gene (LOC_Os08g42020) full-length cDNA structure. (B) Amino acid sequence alignment of Arabidopsis and rice BRP5 proteins. Identical and similar amino acids are highlighted in black and gray, respectively. The red box indicates the positions of the four conserved cysteine bases in the zinc ribbon domain. The green arrows indicate the conserved B-finger domain. The dark blue and light blue boxes indicate the two repeat core domains. (C) A phylogenetic tree of the TFIIB-related proteins from Arabidopsis and rice. A phylogenetic tree was constructed using each gene’s protein sequence with MEGA 6.0 software.

Figure 2

Analysis and subcellular localization of BRP5 expression in rice. (A) Expression activities of transgenic rice OsBRP5-GUS in different tissues and organs. Leaf blade (a); radicle (b); matured root (c); flower (d), glume (e); stamen (f); and pollen grains (g). (B) OsBRP5 subcells were localized in the epidermal cells of N. benthamiana, OsBRP5-GFP fusion, or GFP co-expressed with the nuclear marker H2B-RFP alone, and the signal was observed using a fluorescence microscope. Green fluorescence shows GFP, red fluorescence shows nucleus fluorescence, and yellow fluorescence shows the merged fluorescence. Bars = 20 µm.

Figure 3

OsBRP5 interacted with OsTBP. (A) pGBKT7-OsTBP1/2/3 does not cause autoactivation or toxicity when introduced in yeast. (B) Yeast two-hybrid assays of OsBRP5 proteins with OsTBP3. Prey AD, the Gal4 activation domain fusion; bait BD, the Gal4 DNA-binding domain fusion. AD-T and BD-p53, the positive control; AD-T and BD-Lam, the negative control in the Y2H assay. The indicated fusion bait and prey constructs were co-transformed into yeast cells. The transformed yeast was cultured with SD/-Leu-Trp medium, and SD/-Leu-Trp-His-Ade was selected as the medium.

Figure 4

BiFC analysis of OsBRP5 and OsTBP interaction in N. benthamiana leaves. OsTBP 1/2/3 (OsTBP1/2/3-N-YFP) fused with the n terminal of YFP and OsBRP5 fused with the c terminal of YFP (OsBRP5-N-YFP) were infected and co-expressed in N. benthamiana for 48 h. YFP fluorescence (green), RFP autofluorescence (red), bright field, and combination images were taken using confocal microscopy. Bars = 50 μm.

Figure 5

Targeted mutations of the rice OsBRP5 gene using CRISPR/Cas9 genome editing. (A) Comparison of OsBRP5 CDS in WT plants and five osbrp5 mutants. The red box represents the target site, and the blue color indicates PAM. (B) Identification results of five osbrp5 mutants.

Figure 6

The silencing of OsBRP5 in rice affected pollen development and plant fruit setting. (A) Diagram of the OsBRP5-RNAi construct. Fragments containing the OsBRP5 segment in the sense and antisense orientation separated by an unrelated spacer were driven by the CaMV 35S promoter. (B) The transcription levels of OsBRP5 in WT and OsBRP5R plants. Error bars represent standard deviation (n = 3). (C) The panicles of WT and OsBRP5R plants. The green and empty green seeds are marked by arrows. (D) The mature seeds from wild-type and OsBRP5R plants. The empty seeds in OsBRP5R are highlighted by white arrows. (E) The total grains in the plant of OsBRP5R and WT rice. (F) I₂-KI staining of mature pollen grains of WT and OsBRP5R plants. Bars = 500 μm. (G) The statistics for the pollen viability rate of WT and OsBRP5R plants. Data are shown as means ± SE (n = 3). (H) The statistics for the seed setting rate of WT and OsBRP5R plants. Data are shown as means ± SE (n = 6). Statistically significant differences are indicated by different symbols (** p < 0.01; *** p < 0.001. “ns” indicates no significant difference. one-way ANOVA with statistical significance using Student’s t-test).