A natural allele of OsMS1 responds to temperature changes and confers thermosensitive genic male sterility

Abstract

Changes in ambient temperature influence crop fertility and production. Understanding of how crops sense and respond to temperature is thus crucial for sustainable agriculture. The thermosensitive genic male-sterile (TGMS) lines are widely used for hybrid rice breeding and also provide a good system to investigate the mechanisms underlying temperature sensing and responses in crops. Here, we show that OsMS1 is a histone binding protein, and its natural allele OsMS1^wenmin1 confers thermosensitive male sterility in rice. OsMS1 is primarily localized in nuclei, while OsMS1^wenmin1 is localized in nuclei and cytoplasm. Temperature regulates the abundances of OsMS1 and OsMS1^wenmin1 proteins. The high temperature causes more reduction of OsMS1^wenmin1 than OsMS1 in nuclei. OsMS1 associates with the transcription factor TDR to regulate expression of downstream genes in a temperature-dependent manner. Thus, our findings uncover a thermosensitive mechanism that could be useful for hybrid crop breeding.

Fig. 1. Phenotypes and identification of the wenmin1 allele.

a Pollen fertility of ZJ1, Tian1S and ZJ1-wenmin1 at the indicated day average temperature (DAT). b Statistical analysis of (a). Data are presented as means ± s.e.m. (n = 10 biological replicates). P values indicate the significant differences relative to ZJ1 at the indicated temperatures. Two-tailed unpaired t-test was used for statistical analysis. n.s., not significant. c Mapping and identification of wenmin1. Arrows, predicted genes; black boxes, exons; black lines, introns. Numbers indicate the exon length. The specific single nucleotide polymorphism in Tian1S, HNS-1, and ZJ1-wenmin1 was shown in red. d OsMS1-GUS is specifically detectable in the stages 8 and 9 anthers. S7, Stage 7; S8, stage 8; S9, stage 9; S10, stage 10. e Schematic diagram of OsMS1 protein structure. Green, purple and red rectangles indicate NLS, LXXLL motif, and PHD domain, respectively. aa, amino acids. The mutated amino acids (L301P) in Tian1S, HNS-1, and ZJ1-wenmin1 are shaded in red. f Spikelet (up) and pollen (down) phenotypes of ZJ1, ZJ1-wenmin1, and gOsMS1:ZJ1-wenmin1 lines (#1 and #2). Scale bars, 100 μm (a), 1 mm (d), 1 mm (f, up), 100 μm (f, down). Representative results of at least three independent experiments in (d) and (f) are shown. Source data are provided as a Source data file.

Fig. 2. OsMS1 is a histone binding protein.

a In vitro histone pull-down assay with the indicated proteins against biotinylated peptides. Proteins were pulled down by strepavidin beads and detected by western blots with antibody against FLAG or GST. b OsMS1 associates with H4 in a LCI assay. The indicated constructs were transiently expressed in N. benthamiana at 22 °C for 48 h. Luciferase activity is depicted with false color from low (blue) to high (red). Nluc, N-terminal half of firefly luciferase; Cluc, C-terminal half of firefly luciferase. c OsMS1 associates with H4 in a BiFC assay. Assays were done as in (b). nYFP, N-terminal portion of YFP; cYFP, C-terminal portion of YFP. Black arrows indicate nuclei. Scale bars, 50 μm. d OsMS1 associates with H4 in a semi-in vivo pull-down assay. GST-OsMS1 was incubated with nuclear proteins extracts from stage 9 anthers of rice young panicles. Proteins were pulled down by Glutathione sepharose beads and detected by western blots with antibody against GST or H4. e OsMS1 associates with H4 in a co-immunoprecipitation assay. Proteins extracts from stage 9 anthers of transgenic 35S:GFP-OsMS1 rice was immunoprecipitated with a GFP-Trap^®_A agarose beads and detected by western blots with antibody against GFP or H4. 35S:GFP transgenic rice was used as a negative control. Experiments were repeated more than three times with similar results. Source data are provided as a Source data file.

Fig. 3. Temperature regulates the abundances of OsMS1 and OsMS1^wenmin1.

a Transcriptional activation assay in yeast. Yeast cells were spotted on the indicated plates and grown at low (22 °C) or high (30 °C) temperatures for 96 h. BD, binding domain; AD, activation domain; SD-2, plates lacking Trp and Leu; SD-3 + 2 mM 3-AT, plates lacking Trp, Leu, and His, plus 2 mM 3-aminotriazole. b, c Protein levels of GFP-OsMS1 and GFP-OsMS1^wenmin1 are temperature-dependent in yeast cells (b) and N. benthamiana (c). N. benthamiana were incubated at 22 °C for 44 h and then transferred to the indicated temperatures for another 4 h before detecting the GFP signals. Red arrows indicate nuclei, white arrows indicate cytoplasm. d Quantitative analysis of (c). The GFP-OsMS1 and GFP-OsMS1^wenmin1 protein levels were quantitated and normalized to the mCherry level. P values indicate the significant differences relative to GFP-OsMS1 at 22 °C. e Dual-luciferase system. LUC, firefly luciferase. Ren-LUC, Renilla luciferase. f Relative luciferase activity of OsMS1-LUC and OsMS1^wenmin1-LUC in N. benthamiana plants during the low temperature to high-temperature transition. g OsMS1 and OsMS1^wenmin1 mRNA levels as in (f). h–j Protein levels of OsMS1-GUS and OsMS1^wenmin1-GUS are temperature-dependent in rice. pOsMS1:OsMS1-GUS and pOsMS1:OsMS1^wenmin1-GUS transgenic plants were grown at 22 °C or 30 °C in growth chamber. Protein levels of OsMS1-GUS and OsMS1^wenmin1-GUS were detected by GUS-staining of transgenic plant spikelets at stage 9 (h). OsMS1-GUS (i, left) and OsMS1^wenmin1-GUS (j, left) proteins were detected by ani-GUS antibody. Quantifications of OsMS1-GUS (i, right) and OsMS1^wenmin1-GUS (j, right) protein levels were relative to ACTIN. Scale bars, 5 μm (b), 50 μm (c), 1 mm (h). In d, f, g, i (right) and j (right). Data are presented as means ± s.e.m. (n = 6 biological replicates for d, n = 3 biological replicates for f, g, i, and j). Two-tailed unpaired t-test was used for statistical analysis. n.s., not significant. Source data are provided as a Source data file.

Fig. 4. The interactions of OsMS1 and OsMS1^wenmin1 with TDR in a temperature-dependent manner.

a Temperature-dependent association of OsMS1^wenmin1 and TDR in yeast. Yeast cells (OD₆₀₀ = 1) were spotted on the SD-4 plates and grown at 22 °C, 26 °C, or 30 °C for 96 h. b Temperature-dependent associations of OsMS1 and OsMS1^wenmin1 with TDR in LCI assays in N. benthamiana plants. N. benthamiana plants infiltrated with the Agrobacteria combinations were incubated at 22°C for 44 h and then transferred to the indicated temperatures for another 4 h. c Quantitative analysis of (b). Box-and-whisker plots display the minimum and maximum, the 25th and 75th percentiles (box), and median (center line). RLU, relative luminescence unit. Data are presented as mean RLU ± s.e.m. (n = 12 biological replicates). Two-tailed unpaired t-test was used for statistical analysis. n.s., not significant. d Temperature-dependent associations of OsMS1 and OsMS1^wenmin1 with TDR in BiFC assays in N. benthamiana plants. nYFP, N-terminal portion of YFP; cYFP, C-terminal portion of YFP. Scale bars, 1 μm. e OsMS1 and OsMS1^wenmin1 interact with TDR in a GST pull-down (PD) assay. The proteins were detected by western blots with antibody against GST or FLAG. Experiments in (e) were repeated three times with similar results. Source data are provided as a Source data file.

Fig. 5. OsMS1 enhances the binding of TDR to the EAT1 promoter.

a Schematic representation of partial structure of the EAT1 locus. The conserved E-box elements (CANNTG) are shown in black boxes. The mutated sequences are shown in red. P1 and P2 represent DNA fragments used for ChIP-qPCR analysis. P1 fragment contains two E-box elements, while P2 fragment contains no E-box element. b EMSA assay reveals that TDR but not OsMS1 directly binds to the EAT1 promoter in vitro. The wild-type and mutated probes were shown in (a). MBP was used as a negative control. 10-, 20-,100-fold excess unlabeled probes were used as competitors. c EMSA assay showing the effects of OsMS1 and OsMS1^wenmin1 on the binding of TDR to the E-box elements in the EAT1 promoter in vitro. Biotin-labeled probes were incubated with a constant amount of MBP-TDR protein with increasing amounts of MBP-OsMS1, MBP-OsMS1^wenmin1 or MBP. MBP was used as a negative control. d, e ChIP-qPCR assays showing the effects of OsMS1 and OsMS1^wenmin1 on the binding of TDR to the EAT1 promoter in rice leaf protoplasts (n = 3 biological repeats). Chromatins were incubated with anti-MYC antibody, and precipitated by protein A + G magnetic beads. ChIP-qPCR results were quantified by normalization of the MYC-IP signal with the corresponding input signal (IP/input), and different letters represent significant differences (P < 0.01; Duncan’s multiple range test). f Temperature-dependent enrichment of OsMS1 and OsMS1^wenmin1 at the EAT1 promoter by ChIP assays in rice stage 9 anthers. pOsMS1:OsMS1-FLAG and pOsMS1:OsMS1^wenmin1-FLAG transgenic lines were grown at 22 °C or 30 °C in growth chamber. Chromatins were incubated with anti-FLAG antibody, and precipitated by protein A + G magnetic beads. ChIP-qPCR results were quantified by normalization of the FLAG-IP signal with the corresponding input signal (IP/input). P values indicate the significant differences relative to pOsMS1:OsMS1-FLAG at 22 °C. g mRNA levels of OsMS1-FLAG and OsMS1^wenmin1-FLAG in (f). h qPCR analysis of EAT1 expressions in ZJ1 and ZJ1-wenmin1 stage 9 anthers. ZJ1 and ZJ1-wenmin1 were grown at 22 °C or 30 °C in growth chamber. For d–h, values are means ± s.e.m. (n = 3 biological repeats). For f–h, two-tailed unpaired t-test was used for statistical analysis. n.s., not significant. Experiments in b and c were repeated three times with similar results. Source data are provided as a Source data file.