Anther‐specific Expression of OsRIP1 Causes Dominant Male Sterility in Rice
Dekun Lei,Anqi Jian,Xianbo Huang,Lifeng Xi,Liangming Chen,Weiliang Bai,Siqi Cheng,Xiaodong He,Ye Xiong,Xiaowen Yu,Chaolong Wang,Zheng Hai,San‐Lin You,Qiming Wang,Jiayu Lu,Hu Yang,Zhipeng Xie,Ling Jiang,Xin Zhang,Yulong Ren,Cailin Lei,Zhijun Cheng,Qian Lin,Chuanyin Wu,Shanshan Zhu,Zhigang Zhao,Jianmin Wan
DOI: https://doi.org/10.1111/pbi.14140
IF: 13.8
2023-01-01
Plant Biotechnology Journal
Abstract:Rice (Oryza sativa L.) is a major staple food crop feeding more than half of the world human population. Hybrid rice, using the three-line or two-line strategy for hybrid seed production, produces 10%–20% higher grain yield than the conventional rice and thus has been widely commercialized (Cheng et al., 2007). A number of genes are involved in the control of cytoplasmic male sterility (CMS) or photoperiod/thermos-sensitive genic male sterility (P/TMS), have been reported and successfully used in hybrid rice breeding (Chen and Liu, 2014). The Sanming-Dms line (named Dms1 hereafter) has been widely used for recurrent selection in breeding programmes in China (Zhang et al., 2022). Use of Dms1 can eliminate the hand emasculation procedure, thus enabling large-scale crossing with various parents. At each selection, only the male-sterile plants with traits of interest are chosen as female to cross with other breeding materials. The male sterility is controlled by a single dominant locus, and has been mapped on the short arm of chromosome 8 (Yang et al., 2012), however, the underlying gene has not been cloned. To clone Dms1, we constructed a near-isogenic line (NIL)-NJ4Dms1 contains the genetic background of the japonica rice variety Ningjing4 (NJ4) but carries a chromosome segment insertion with Dms1 introduced from the mutant donor (Figure S1). Due to the dominant nature of male sterility in Dms1, NIL-NJ4Dms1 must be maintained as heterozygote at the Dms1 locus. Compared to NJ4, anthers of NIL-NJ4Dms1 are much smaller (Figure 1a). In contrast, NIL-NipDms1 set seeds at a rate close to the recipient Nipponbare (Nip) when pollinated with wild-type pollens, indicating the Dms1 locus has little effect on female gamete fertility (Figure S2). Further, iodine-potassium iodide staining assay revealed completely aborted pollens in NIL-NJ4Dms1, again verifying the male specificity of the Dms1-associated reproductive sterility (Figure S3). No obvious morphological difference was observed between NJ4 and NIL-NJ4Dms1 (Figure S3). Acetocarmine staining showed defective meiosis of the pollen mother cell in NIL-NJ4Dms1 (Figure S4). Detailed cytological observations uncovered that the pollen mother cell (PMC) in NIL-NJ4Dms1 failed to complete meiosis, resulting in collapsed anthers at the S7 and S8 stages. Ruptured cell membrane and vacuoles were present in PMC, a phenomenon similar to necroptosis in mammal cells suffering from toxin or other harmful external environments (Figure 1b and Figure S5). Thus, the pollen-less anther is the cause of male sterility in NIL-NJ4Dms1. To fine-map Dms1, an F2 population with 934 individuals derived from the cross of NIL-NJ4Dms1 with NJ4 was developed. Dms1 was primarily located to a 290-kb interval between the molecular markers xhy45 and xhy54 on chromosome 8, and subsequently fine-mapped to a 5.4-kb region between the markers xhy128 and xhy129 using 19 348 plants (Figure 1c). This region contains a putative gene, LOC_Os08g03820, encodes a 32.5-kDa ribosome-inactivating protein (RIP), thus named OsRIP1 herein. Sequence comparison found a 1978-bp insertion 98 bp upstream of the start codon of OsRIP1 in NIL-NJ4Dms1 (Table S1). qRT-PCR analysis showed over threefold higher OsRIP1 expression in young inflorescences of NIL-NJ4Dms1 than in NJ4 (Figure 1d). We cloned the promoter region from Nip and NIL-NipDms1, respectively, and used them to drive expression of the GUS reporter gene. Intriguingly, GUS expression was seen in the pOsRIP1Dms1::GUS plants, but not in the pOsRIP1Nip::GUS plants (Figure 1e). In situ hybridization to anthers at stages 8a and 8b further confirmed OsRIP1 expression in NIL-NipDms1 but not in Nip (Figure 1e and Figure S6). Next, we knocked out OsRIP1 in Nip and NIL-NipDms1, respectively, using the CRISPR/Cas9 technology. We observed fertility restoration in NIL-NipDms1 knockout plants, while no fertility change was seen in Nip knockouts (Figure 1f). When using the anther-specific EAT1 promoter to drive OsRIP1 in Nip, the male sterility phenotype mimicking NIL-NipDms1 appeared (Figure 1g). Moreover, manipulated root- or leaf-specific expression of OsRIP1 resulted in lethal plants (Figure S7), which demonstrate that OsRIP1 has the conserved toxicity to rice cells. Taken together, our results demonstrate that LOC_Os08g03820 is responsible for Dms1 and the enhanced expression by the insertion that causes male sterility. The mechanism of RIP producing cytotoxicity in both mammal and plant are elusive (Grela et al., 2019). It is known that RIPs and homologous modelling of protein structure reveals five amino acids that aggregate to form active centres (Figure 1h). To check whether OsRIP1 has the similar enzymatic activity, the HIS-OsRIP1 and HIS-mutants (substitution at the five sites: HIS-OsRIP1Y100A, HIS-OsRIP1Y146A, HIS-OsRIP1E198A, HIS-OsRIP1R201A and HIS-OsRIP1W236E) recombination proteins were expressed in E. coli. The HIS-OsRIP1, but not the other five mutant proteins, showed significant 28SrRNA depurinate activity in rabbit reticulocyte lysate (Figure 1i), suggesting that OsRIP1 is a typical ribosome-inactivating protein. Conversely, neither HIS-OsRIP1 nor its mutant versions had depurinate activity in wheat germ extract (Figure 1j). Surprisingly, these five non-enzyme-active mutant proteins still caused cell death after introduced into tobacco (Figure 1k). To investigate how OsRIP1 produces cytotoxicity in plants, semi-in vivo immunoprecipitation–mass spectrometry was performed. GO analysis (gene ontology) indicated that proteins involved in structure constituent of ribosome and translation process were enriched (Figure 1l), implying that OsRIP1 might produce cytotoxicity by inhibiting ribosome translation function. To test this possibility, we performed in vitro translation inhibition experiments using cell-free protein synthesis system. As anticipated, OsRIP1 and all the five mutant versions showed strongly inhibitory effects on translation in mammal and plant (Figure 1m,n). All those data suggest that OsRIP1 toxifies plant cells by inhibiting translation function rather than RNA depurating activity. In China, the Taigu-dominant male-sterile gene Ms2 has been widely used for population improvement by recurrent selection in wheat conventional breeding (Ni et al., 2017; Xia et al., 2017). However, such an approach has not been possible in rice. In this study, we have cloned the Dms1 locus and demonstrated why the male sterility is dominant in the Dms1. The use of dominant male sterility (Dms) lines for recurrent selection can at least save one generation of selfing, compared to recessive male sterility lines. Thus, isolation of Dms causal genes will facilitate creation of Dms in other varieties. Male sterility in Dms1 is a much-preferred line in rice breeding. The Dms1 line is an ideal line widely used in the recurrent selection breeding strategy. Identification of Dms1 provides opportunity to develop other dominant male sterility lines in various genetic backgrounds by transformation approach, which will increase efficiency and reduce cost in breeding programmes. Recent studies have discovered a number of genes involved in male gametogenesis, and their malfunctionization leads to male sterility. It is possible that the activated gametogenesis-specific expression of OsRIP1 directly or indirectly affects translation of some, if not all, of those genes, resulting in defective gametogenesis. Mammals suffer from toxicity when accidently foraging plants containing RIPs, and the uptaken RIPs in mammals specifically interact with stalk subunit of ribosome, then depurinate 28SrRNA (Grela et al., 2019). The damaged ribosome is no longer a functional organelle for normal protein translation. However, RIPs exist in plants, but cannot depurinate plant ribosomes and its toxicity to plants is unclear (Nielsen and Boston, 2001; De Zaeytijd and Van Damme, 2017). Our results also indicate that OsRIP1 inhibits translation function of ribosome not by 28SrRNA depurination. As the cytotoxicity of OsRIP1 is likely conserved in monocot and dicot plants, we anticipate OsRIP1 can be employed to create new Dms lines, for example, by anther-specific expression in other crops. This work was supported by the National Natural Science Foundation of China (31991224, U2002202 and 31971909), the National Key Research and Development Program of China (2022YFD1201504 and 2022YFF1003503), the Key Research and Development Program of Jiangsu Province (BE2021360) and the Natural Science Foundation of Jiangsu Province (Grant No. BK20200023). The authors declare no conflict of interest. Jianmin Wan, Zhigang Zhao and Shanshan Zhu supervised the project. Dekun Lei and Anqi Jian performed the experiments. Xianbo Huang provided the Dms material. Other authors provided technical supports. Dekun Lei, Anqi Jian prepared the article. Figure S1-S7 Supplementary Figures. Table S1-S2 Supplementary Tables. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.