Fine Mapping And Clone Of A New Hybrid Embryo Sac Abortion Gene, S31(t), In Rice (Oryza Sativa L.)

This S5ⁿ has been incorporated into indica or japonica variety to overcome sterility problem in wide crosses and hybrid rice breeding. So far, the S5ⁿ allele has been effective to produce fertile hybrids in a large number crosses. In the analyses of many indica-japonica hybrids, it has also been observed that there is considerable variation in the fertility level in hybrids from the same WCV crossed to different indica or japonica varieties. Thus, hybrid sterility in different varietal combinations may be explained by allelic interaction at different loci.A japonica cultivar, USSR₅ from former Soviet Union is identified to show strong genninability under cold stress, and Guangjie 9 is an indica cultivar from Guangdong province, China, with weak genninability under cold stress. The hybrid of USSR₅/Guangjie 9 reveals hybrid sterility. Therefore, we performed a genome-wide analysis by assaying a backcross population of USSRs/Guangjie 9//USSR₅ to determine the number and genomic location of genes controlling the hybrid sterility. As the result, two loci conferring the hybrid fertility were found, which acted independently of each other. Interestingly, the locus on chromosome 4 coincided with the previously identified S9, while the locus on chromosome 5 was distinct from all the previously reported ones. The new locus was thus designated as S31(t). Based on allelic interaction causing female gamete abortion, two alleles were found at S31(t), i.e. S31(t)-un in USSR₅ and S31(t)-gi in Guangjie 9. An Aus variety, Dular, was assumed to have a neutral allele, S31(t)-n. The SSR markers of RM5586 and RM13 were considered to be useful for marker-aided transfer of the wide compatibility genes in the hybrid rice breeding.To further understand the genetic mechanism of the hybrid sterility, in this study, two sets of F₁S were constructed using Asominori, a typical japonica variety, and 02428, a wide-compatibility variety, as the pollen parents, and the 66 CSSLs as the maternal parents, respectively. Typical spikelet semi-sterility was observed in F₁ hybrids between Asominori and CSSL34 which was previously reported to have a sterility gene of S31(t). The embryo sac fertility, pollen fertility and spikelet fertility of three parents CSSL34, 02428, and Asominori were normal, while spikelet sterility was observed in F₁ hybrid between CSSL34 and Asominori, and the reciprocal F₁ hybrid did not significantly differ in their spikelet fertility, which indicating no cytoplasmic effect on the spikelet fertility in the F₁ hybrids. To elucidate the mechanism of the sterility in F₁ hybrids, pollen fertility, in vitro pollen germination, embryo sac fertility, fertilized ovaries, and spikelet fertility were examined and compared to control parent strains. The results revealed that both pollen fertility and in vitro pollen germination rate of the F₁ hybrids were similar to those of the parents. Besides, pollen and spikelet fertilities in the BC1F1 of Asominori//CSSL34/Asominori were normal. In addition, the spikelet fertility of the F₁ hybrids of CSSL34/Asominori was not restored to normal levels after hand pollination with pollen from each parent. Furthermore, the number of pollen grains adhered to stigmas, pollen germination, and pollen tube elongation has no distinction between parents and F₁ hybrids by inspection with confocal laser scanning microscopy. On the other hand, the frequency of abnormal embryo sacs in the CSSL34/Asominori hybrid was significantly higher than in parents. The embryo sac plays a pivotal role in sexual reproduction of angiosperms. In order to study the stage of abnormal embryo sac, differences in embryo sac development between the CSSL34/Asominori hybrid and parent were analyzed using series sections. However, in some of the ovaries from the CSSL34/Asominori hybrid plant, megamametogenesis appeared to be blocked at the first round of mitosis. The degenerating functional megaspore is characterized by no vacuole differentiation, less or withered cytoplasm. Taken together, these results suggested that the partial abortive embryo sac was the main reason for the hybrid sterility in the F1 hybrids between CSSL34 and Asominori.A population of 1630 F₁ plants derived from the three-way cross CSSL34/02428//Asominori was developed to further fine-map S31(t). Based on the physical location of molecular markers, S31(t) was finally delimited to a region of 54-kb between InDel193 and InDel223. Sequence analysis of this fragment revealed eight predicted open reading frames, six of which encoded known proteins and two encoded putative proteins. These results are useful in map-based cloning of S31(t) and marker-assisted transferring of the neutral allele in rice breeding programs.The spikelet fertility under high temperature and heat susceptibility index (HSI) of spikelet fertility were used to evaluate the tolerance of rice to heat stress at booting stage, respectively. Two QTLs related to relative heat tolerance at the booting stage were detected, with LOD scoring 3.81 and 2.86 on chromosome 4 and 8, and explained 16.8% and 9.9% of the phenotypic variance, respectively. Further analysis of epistatic effected QTLs for relative heat tolerance showed that four pairs of loci exhibited interaction on 8 chromosomes, which indicated that the heat tolerance of spikelet fertility was not only controlled by main effect QTL but also influenced by gene interaction.