| Rice, one of the world's most important crops, is the staple food of over half of the world's population. Nitrogen, one of the macronutrients necessary for plant growth, is an important factor limiting the crops production. Excessive application of nitrogen fertilizer not only causes serious environmental pollution, but also increases the rice production cost. Rice has rich and diverse genetic resources in the natural population. Association mapping, which based on linkage disequilibrium (LD) and used natural population as the basic research material, has become a more and more popular method to discover and utilize important genes in rice. Association mapping is not only an effective approach to discover new genes, but also a bridge to link the structural genomics and phenomics. It could directly identify the alleles that associated with the target traits. With the rapid development of molecular biology and DNA sequencing technology, lots of genes have been cloned and functional characterized. The objectives of this study are (1) to investigate the nucleotide diversity of nitrogen metabolism genes in a rice mini-core collection and a panel of O. rufipogon accessions using sequencing technology, (2) to clarify the LD levels within and between different genes, and (3) to demonstrate the relationship of polymorphism sites and yield-related traits under normal and low nitrogen conditions. The alleles explored in this study would provide useful guides in rice breeding of nitrogen efficiency. In addition, this study is also aim to reveal origin and evolution of OsAMT1;1 in rice. The main results are as follows:1. We obtained the sequences of seven nitrogen metabolism genes (OsAMT1;1, OsNRT1, OsNR1, OsGS1;1, OsNADH-GOGAT1, OsAS1 and OsGDHl) in a panel of 216 rice accessions (including indica, japonica and its wild relatives) using the sequencing technology of PCR products. The alignment of the genes ranged from 1608 bp (OsNADH-GOGAT1) to 3506 bp(OsGDH1), with a total length of 16,899 bp. A total of 72 insertion/deletions (Indels) were identified exclusively in introns and 5' or 3' noncoding regions. Totally 330 single nucleotide polymorphisms (SNPs) were found across the 216 rice accessions, with the SNP number being 97 in indica,71 in japonica, 103 in O. sativa and 240 in O. rufipogon.2. The nucleotide diversity of the seven nitrogen metabolism genes was compared at the taxon level, including indica, japonica, O. sativa and O. rufipogon. Overall, the nucleotide diversity level of O. rufipogon was significantly higher than that in indica and japonica, and indica and japonica approximately maintained 33.5% and 23.4% as much diversity as O. rufipogon, respectively.3. The nucleotide diversity was also inverstigated in O. rufipogon, landraces and elite cultivars, and the results showed that about half (45.7%) of the nucleotide diversity of the nitrogen metabolism genes was lost during the transition from O. rufipogon to landraces while it was nearly unchanged during the transition from landraces to elite cultivars.4. LD plots were evaluated for six nitrogen metabolism genes (without OsAMT1;1) in 190 O. sativa, and different levels of LD were revealed. OsGDHl exhibited the highest LD level and the average value of r2 was 0.96. OsAS1 (average r2=0.62), OsNRTl (average r2=0.45) and OsNADH-GOGAT1 (average r2=0.44) showed medium LD levels, while OsNRl (average r2=0.35) and OsGSl;1 (average r2=0.28) showed the lowest LD levels. In addition, high LD levels were also observed among five genes of OsNR1, OsNRTl, OsGS1;1, OsNADH-GOGATl and OsGDH1.5. Totally twenty-four main haplotypes (n> 5) were observed for seven nitrogen metabolism genes in 190 O. sativa, the number of haplotypes ranged from one to five with an average of 3.4 per gene. In addition, the haplotypes (except haplotype H1) were indica-like or japonica-like. The results of haplotype analysis showed that in each sub-pathway, indica-like and japonica-like haplotypes were always linked with themselves-like haplotypes for strong LD levels, although these nitrogen metabolism genes were located on different chromosomes. This may be because indica and japonica have the different geographic origins.6. Based on general linear model considering population structure, association mapping was conducted between polymorphisms (or haplotypes) and the yield-related traits in seven environments. Overall, OsNRT1 associated with YPP and SS, OsNRl associated with TGW, OsGS1;1 associated with SS and YPP, OsNADH-GOGATl associated with PNPP and TGW, OsASl associated with PNPP, GNPP and YPP, and OsGDH1 associated with TGW. In addition, the favorable allele (or haplotype) of each nitrogen metabolism gene was identified, and the effects of some alleles (or haplotypes) were validated by using near isogenetic lines. The information explored in this study could be applied in the rice breeding programs.7. The nucleotide polymorphism (Ï€) of OsAMT1; 1 in O. sativa (including indica and japonica) was as low as 2.3% of that in O. rufipogon, and the values of Tajima's D and Fu and Li's D* were both negative significant in O. sativa, suggesting OsAMT1;1 was under positive selection. Most (182/190) accessions of O. sativa have the same OsAMT1;1 allele. Compared with two other genes (Loc_Os01g48960 and Loc_Os02g50240) that were used as the control, the nucleotide diversity of OsAMT1;1 was severely reduced in landraces and elite cultivars.8. The nucleotide diversity of additional seven DNA fragments surrounding OsAMT1;1 was inverstigated in 94 O. sativa and 19 O. rufipogon, and the results showed that the nucleotide diversity of O. sativa was severely reduced in the OsAMT1;1 genomic region spanning approximately 100 kb long compared with O. rufipogon. EHH (Extended Haplotype Homozygosity) analysis showed that LD remained high levels across the 100 kb genomic region around OsAMT1;1 in O. sativa, but fell rapidly in O. rufipogon. All the results suggest that OsAMT1;1 was under strong selection. |