| Early maturity and fast grain filling rate are two main characteristics of the wheat varieties in China, and grain filling rate is one of the crucial factors affecting the final grain yield of wheat in most of the Chinese wheat regions. Few genetic researches were carried out on the grain filling rate for its complex experimental procedure though it was proved to be controlled by genetic factors. To understand the genetic basis of grain filling rate and related traits such as thousand grain weight, grain length, grain width, grain thickness, grain volume, maturity period and plant shape, quantitative trait locus (QTL) analysis was conducted using phenotypic data from a population of 142 recombinant inbred lines (RIL) derived from a cross between Yu8679 and Heshangmai across four environments Beijing 2006, Beijing 2007, Chengdu 2007 and Hefei 2007. The main results were as follows:1. The dynamics of grain filling showed"S"curve with the characteristics of"slow-fast-slow". During the grain filling duration, all the lines had a fast filling stage but the range of grain increase, the time and the days of fast filling stage were different; The correlation analysis showed that significant positive correlations were present between mean grain filling rate, maximum grain filling rate, grain length, grain width, grain thickness, grain volume and grain weight per ear. The correlation between grain volume and thousand grain weight was the highest (r =0.931, P < 0.0001), followed by the one between mean grain filling rate and thousand grain weight (r =0.85, P < 0.0001). The path-coefficient analysis for thousand grain weight showed that the coefficients of direct path between grain volume and thousand grain weight was the highest, which was 0.636, followed by grain weight per ear (0.338) and mean grain filling rate (0.089). The regression equation for grain volume, grain weight per ear and mean grain filling rate to thousand grain weight was Y=-13.117+8.768GV+6.000Gwe+3.339GFRmean.2. Out of the 1129 SSRs and ten ESTs-markers detected, 186 (184 SSRs and 2 ESTs-markers) turned out to be polymorphic between the parents, and the ratio of polymorphism was 16.47%. Finally, a total of 170 SSRs and two ESTs-markers (Tx23-24 and Tx37-38) were located on all the chromosomes of wheat except chromosome 6A. The linkage map reveals a total length of 1584.6 cM with an average interval length of 9.32 cM, the number of loci per linkage group ranges from 3 (chromosome 3A) to 18 (chromosome 2A) with a mean of 8.6 loci per chromosome.3. The QTL analysis of thirteen agronomic traits including mean grain filling rate (GFRmean), maximum grain filling rate (GFRmax), thousand grain weight (TGW), grain filling duration (GFD), grain length (GL), grain width (GW), grain thickness (GT), grain volume (GV), plant height (PHT), grain numbers per ear (Gne), grain weight per ear (Gwe), flowering time (FT) and maturation time (MT) were performed with the software QTL Cartographer 2.5 based on composite interval mapping (CIM) under four environments (Beijing 2006, Beijing 2007, Chengdu 2007 and Hefei 2007). A total of 172 putative QTL were identified for all the traits on all the wheat chromosomes except chromosome 6A.4. A total of 17 QTL were identified for mean grain filling rate over four environments, accounting for phenotypic variations of GFRmean by 7.17%~20.83%. A total of 16 QTL were identified for maximum grain filling rate, accounting for phenotypic variations of GFRmax by 6.46%~15.95%. A total of 21 QTL were identified for thousand grain weight, accounting for phenotypic variations of TGW by 4.36%~16.8%. A total of 6 QTL were identified for grain filling duration, accounting for phenotypic variations of GFD by 6.68%~15.72%. A total of 17 QTL were identified for grain length, accounting for phenotypic variations of GL by 6.26%~19.02%. A total of 16 QTL were identified for grain width, accounting for phenotypic variations of GW by 6.17%~29.87%. A total of 18 QTL were identified for grain thickness, accounting for phenotypic variations of GT by 6.60%~17.35%. A total of 21 QTL were identified for grain volume, accounting for phenotypic variations of GV by 5.18%~24.24%. A total of 8 QTL were identified for flowering time, accounting for phenotypic variations of FT by 7.65%~15.77%. A total of 7 QTL were identified for maturation time, accounting for phenotypic variations of MT by 7.14%~10.77%. A total of 6 QTL were identified for plant height, accounting for phenotypic variations of PHT by 5.83%~25.24%. A total of 8 QTL were identified for grain numbers per ear, accounting for phenotypic variations of Gne by 6.87%~15.82%. A total of 10 QTL were identified for grain weight per ear, accounting for phenotypic variations of Gwe by 5.93%~24.06%.5. Many genomic regions were found affecting more than one trait, which indicated the significant correlation between grain filling rate and yield-related traits, especially on chromosomes 1B, 2A and 3B. For example, QTgw.nfcri-1B, QTgw.nfcri-2A and QTgw.nfcri-3B proved to be simultaneously associated with several traits such as thousand grain weight, mean grain filling rate, maximum grain filling rate, grain size, grain volume and grain weight per ear, and they could be detected in three or more than three environments and could explain high phenotypic variation. These three genomic regions with plenty of QTL may play important roles in the development of grain yield. Moreover, the three genomic regions have been confirmed by the single marker regression analysis, focusing on TGW and GFRmean, the alternative allele effects analysis indicated that the more positive alleles of xwmc419 (1B), xgwm359 (2A) and xbarc113 (3B) contributed by the parental line Y8679 the higher TGW and GFRmean. Furthermore, there were significant difference for TGW and GFRmean between the RIL possessing two positive alleles (AAa, AaA, aAA) and one allele (Aaa, aAa, aaA), all those indicated that the three makers were effective and useful for MAS high yield breeding.
|
PDF Full Text Request