Map-based Cloning and Molecular Characterization of the Seed Dormancy 10 Locus in Rice (Oryza sativa L.)

摘要

Seed dormancy is a key adaptive trait of both ecological and agricultural importance. Although many quantitative trait loci (QTL) have been associated with seed dormancy in cereal crops or model plants, a majority of them remain unknown for molecular identities and functions. Cereal crops and wild/weedy relatives usually have weak and strong seed dormancy, respectively, such as Asian cultivated rice vs. weedy rice (Oryza sativa L.). Previous research identified a set of seed dormancy QTL, including qSD10, in single plant-derived BC1F2 and BC1F3 populations from the backcross between the weedy rice line SS18-2 and the cultivated rice line EM93-1. This dissertation aimed to clone qSD10 and characterize the molecular functions of the QTL underlying gene(s). The first objective was to isolate qSD10 as a Mendelian factor and narrow its size by fine mapping to identify candidate genes. A marker-assisted single-plant-selection technique was used to advance a BC1F3 plant for three generations to synchronize the genetic background of qSD10. A population of about 4000 BC1F6 plants segregating for the qSD10-containing region was used to identify recombinants between markers on a partial high-resolution map of 3 mega bases (Mb). The recombinants were selected for marker-assisted progeny testing to delimit qSD10. Finally, qSD10 was narrowed to a genomic region of about 100 kilo bases (Kb). Physiological analysis based on isogenic lines for the 100-kb region demonstrated that qSD10 is involved in the development of primary dormancy by regulating dehydration and acquisition of desiccation tolerance during seed maturation. The progeny testing also identified a QTL for flowering time (qFT10 ) in the same 100-Kb region, with the qSD10/qFT10 haplotype from EM93-1 enhancing seed dormancy and promoting flowering. There are 22 predicted genes in the 100-Kb region, including Os10g32600 annotated as a Myb family transcription factor and previously reported as the Early heading date 1 (Ehd1) gene.;The second objective of this dissertation was to clone the candidate gene Os10g32600 and determine its role in seed dormancy and flowering. Both genomic DNAs (gDNAs) and full-length complementary DNAs (FlcDNAs) of Os10g32600 were sequenced from EM93-1 and SS18-2. The sequence analyses revealed that both Os10g32600 alleles consist of five exons and contain a regulatory domain and a helix-turn-helix DNA-binding domain, suggesting that it could be a Myb family transcription factor (TF) gene. There are seven point mutations that differentiate both alleles, one of which is located in the DNA binding domain. An RNA interference (RNAi) approach was used for functional analysis of the candidate gene. A cDNA sequence of 317 base pairs (bp) was used to design an inverted repeat sequence (IRS) to develop an Os10g32600-RNAi construct. The construct was used to transform the japonica cv. Nipponbare. Genetic analysis for a single copy of the Os10g32600-RNAi transgene in the T2 and T3 generations revealed that silencing Os10g32600 delayed flowering and enhanced seed dormancy. These results revealed that Os10g32600 is the underlying gene of qFT10, and has a pleiotropic effect on seed dormancy in the Nipponbare background. To determine the silencing effects in the EM93-1 background, a Nipponbare Os10g32600-RNAi transgenic line was crossed with EM93-1 to develop BC1F1 and BC2F1 populations. Interestingly, silencing of Os10g32600 delayed flowering in both BC1F1 and BC2F1 populations. However, silencing of Os10g32600 enhanced the dormancy in the BC1F1 but reduced the dormancy in the BC2F1 population. These results suggest that Os10g32600 could be the underlying gene for both qSD10 and qFT10, but its effect on seed dormancy could be reversed by an unknown factor in the genetic background of Nipponbare. The other possibility could be that there is an unknown seed dormancy gene tightly linked to Os10g32600 in the narrowed qSD10 region.;In summary, this research isolated qSD10 as a Mendelian factor and discovered that qSD10 collocates with qFT10 in weedy/wild rice. qFT10 is underpinned by a Myb family transcription factor gene, which also has a pleiotropic effect on seed dormancy. The effect of this Myb gene on seed dormancy is strongly influenced by unknown factors in the genetic background. In addition, qSD10 could be underlain by a gene tightly linked to qFT10 . It is likely that the qSD10/qFT10 haplotype was selected for early flowering in weedy/wild plants to adapt to hot, humid environments in tropical Asia. The narrowed qSD10/ qFT10 haplotype can be used to improve early flowering varieties for resistance to pre-harvest sprouting in rice.