| Rice grain quality improvement has become one of the most important objectives in rice breeding program. Rice grain quality was mainly defined by milling quality, appearance quality, cooking and eating quality and nutritional quality. The complicated genetic basis of grain quality makes it difficult for breeders to improve grain quality efficiently using conventional methods. In present study, the genetic basis of grain quality traits was dissected and the QTL profiles were compared across populations. The major results were as follows:1. Two genetic linkage maps have been constructed. One linkage map was constructed from the population derived from Zhenshan 97B and Delong 208, consisting of 179 SSR markers spanning a total of 1,864.5 cM, with an average interval of 12.2 cM between adjacent markers; and the other was constructed from the population derived from Zhenshan 97B and Nanyangzhan, consisting of 190 SSR markers spanning a total of 1,362.1 cM, with an average interval of 7.8 cM between adjacent markers. There were 65 common markers between two populations with marker orders on chromosomes according well with that in previous report.2. The marker segregation distortion has been analysis using seven populations including two double haploid (DH) populations (Zhenshan 97B/H94 and Zhenshan 87B/Wuyujing 2), and five recombinant inbred (RI) populations (Zhenshan 97B/Delong 208, zhenshan 97B/Nanyangzhan, Zhenshan 97B/Hr5, Zhenshan 97B/Minghui 63 and Zhenshan 97B/IRAT109). The distorted markers were distributed on all 12 chromosomes and most of them linked with each other, forming the segregation distortion regions (SDRs). Totally, 113 non-redundant regions or loci, here it was called as segregation distortion units (SDUs), were identified after the combination of the all results of seven populations. Among them, 53 could be found in at least two populations, and thus called as hot segregation distortion regions (HSDRs). There were two different points in segregation distortion between populations of DH and RIL. First, the RIL populations have different degree of segregation level, while the two DH populations were both in the serious rank. Second, significant skewing in favor of Zhenshan 97B was observed in RIL populations except Minghui 63 population, while the DH populations has a balancing allele composition. These two points might indicate that the DH type has more and complicated distortion-causing factors. The coincident of sterility-related loci with the segregation regions could be one of the explanations for segregation distortion. Much attention should be paid to the regions which were detected in special types of populations or were specific in location, in degree or in direction.3. Twenty-four QTL were identified for the heading date and plant height in rice using three populations, namely, Zhenshan97/Delong208, Zhenshan97/Nanyangzhan, and Zhenshan97/Wuyujing2. Six QTL were detected in at least two populations, while the other 18 were only detected in one of the populations. The positive correlations between the heading date and plant height were found in all populations, and the seven QTL clusters (four after combination) were detected for both traits, with the allele effect in the same direction, indicating the community in their genetic basis. However, these two traits also have their own QTL profiles. Thus, the characterization of these QTL will benefit the breeding.4. The genetic basis of grain length (GL), grain width (GW), grain shape (GS), grain thickness (GT) and one thousand grains weight (TGW) were dissected based on three populations. Both GS and TGW were highly and significantly correlated with GL, GW and GT. However, the correlation was much high with GW in Minghui 63 population, whereas much high with GL in populations Delong 208 and Nanyangzhan. The co-localization of QTL was found for all these traits. The QTL clusters associated with GS or TGW also influenced at least one of the traits of GL, GW and GT. Although basically the GL and GW were controlled by different QTL, the finding of three major QTL clusters for both GL and GW could be a good explanation for the negative correlation between these two traits. The QTL profiles for GT were quite different between different populations, might indicating the less selection on this trait.5. A total of 36 non-redundant QTL associated with the six traits of chalkiness were identified using five populations. Many QTL were identified for both rate and area, and the allele effects from the same parent were in the same directions, indicating a commonality in their genetic bases. The QTL profiles for different types of chalkiness, white belly and white core, were quite different, and so as to their interaction with two environments. This difference implied that the breeding strategy must be made according to the types of chalkiness and the environments. The advantages of the trait partition in genetic studies and breeding programs were also discussed.6. The genetic basis of the cooking and eating quality of rice as reflected by 17 traits (or parameters) was investigated using the population Zhenshan 97/Delong 208. All the traits, except peak paste viscosity (PKV), time needed from gelatinization to peak (BAtime), and cooked rice elongation (CRE), can be divided into two classes. The first class consists of amylose content (AC), gel consistency (GC), and most of the paste viscosity parameters that form a major determinant of eating quality. The second class includes alkali spreading value (ASV), pasting temperature (Atemp) and pasting time (Atime), which characterize cooking process. We identified 26 QTL in two years; nine QTL clusters emerged. The two major clusters, which correspond to the Wx and Alk loci, mainly control the traits in the first and second classes, respectively. Some QTL are co-located for the traits belonging to the same class and also for the traits to a different class. The Wx locus also affects on ASV while the Alk locus also makes minor contributions to GC and some paste viscosity parameters. QTL for CRE and PKV are dispersed and independent of the Wx locus. Low paste viscosity corresponds to low AC and soft gel, which represents good eating quality; high ASV and low Atemp, together with reduced time to gelatinization and PKV, indicate preferred cooking quality. It was also found that the parameters in hearting period, including Atemp, Atime, BAtime, peak temperature, peak time were informative, and the effects of Wx and Alk loci varied in different periods.7. We identified 18 QTL for amino acid profile in milled rice as characterized by the contents of each amino acid, the total contents of essential amino acids, and the total contents of all amino acids. Most QTL co-localized, forming 10 QTL clusters in 2002 and six in 2004, with two major ones detected in both two years. A wide coincidence was found between the QTL detected here and the loci involved in amino acid metabolism pathways including N assimilation and transfer, and amino acid or protein biosynthesis. The results will be useful for identifying candidate genes and marker-assisted transference of the favorable allele in rice breeding programs.
|
PDF Full Text Request