| In this study, Shuhui527 and Minghui86, two widely used elite indica restorers, were used as recurrent parents, and 2004 as donors for developing two BC2F2:5 advanced backcrossing populations, which contain 233 and 226 lines respectively. SAS and One-way ANOVA software were used to analysis phenotypic data and to detect QTL for plant height, panicle length, panicle number per plant, thousand grain weights, spikelet fertility and yield per plant of two populations under irrigation and drought conditions respectively. The main results were as follows:1. Phenotypic performance of advanced backcrossing introgression lines under irrigation and drought conditionsAccording to the performance of mean and ratio of mean for one trait under two environments, plant height (PH), panicle number per plant (PN) and grain yield per plant (GY) were more susceptible to drought than panicle length (PL), thousand grain weight (TGW) and spikelet fertility (SF). From the perspectives of coefficients of variation, PH, PL and TGW were more stable than PN and GY under two environments respectively, while SF was affected by genetic backgrounds.2.Correlation and Path analysis under different backgrounds and different environmentsCorrelation and Path analysis indicated that PN and SF were not only extremely positive and significant correlated with GY, but also could provide the most direct effects for GY under two environments. However, the correlation between PH, PL or TGW and GY varied with the change of genetic backgrounds and environments due to the effect of indirect effects. In addition, extremely positive and significant correlation between different performances of the same trait under two environments was observed for all traits under different backgrounds. The coefficients of PN and GY were quite smaller than others, and 0.29 and 0.27, and 0.28 and 0.22 respectively in the two populations of Minghui86/2004 and Shuhui527/2004. The coefficients of SF in Minghui86/2004 was quite bigger than SF in Shuhui527/2004, and 0.7 and 0.42 respectively. The coefficients of TGW were biggest in two populations, and 0.85 and 0.88 respectively.3. Analysis of QTL mapping under irrigation and drought conditions using Minghui86/2004 populationsA total of 36 QTLs actually were detected under two environments in Minghui86/2004 population, which can be grouped into three types based on their behaviors. Typeâ… included 6 loci which were detected only under irrigation condition; typeâ…¡included 4 loci which were observed only under drought condition; and typeâ…¢consisted of 26 loci which were mapped under two conditions. At the same time, 12 tightly linked or pleiotropic QTLs were also found. By comparison with QTL detection in last generation, some important QTLs with larger effects were extracted, which could be detected in the two generations. Qtgw1c, qtgw3b and qtgw6a, which favorable alleles came from the donor parent, could explain 10.32% and 7.22%, 10.69% and 8.47%, 10.29% and 10.50% of phenotypic variation of TGW under two conditions respectively; and qph1b, which favorable alleles came from the recurrent parent, explained 8.19% and 6.74% of phenotypic variation of PH under two conditions respectively. In this generation, a major loci ie qsf8 was also observed, which favorable alleles was from the recurrent parent, explained 19.59% and 19.55% of phenotypic variation of SF under two conditions respectively.4. Analysis of QTL mapping under irrigation and drought conditions using Shuhui527/2004 populationsA total of 41 QTLs actually were detected under two environments in Shuhui527/2004 population, including 2 for typeâ… , 7 for typeâ…¡, 32 for typeâ…¢and 19 for tightly linked or pleiotropic QTLs. By comparison with QTL detection in last generation, some important QTLs with larger effects were extracted, which could be detected in the two generations. Qtgw1, which favorable alleles was from the donor parent, could explain 10.80% and 16.31% of phenotypic variation of TGW under two conditions respectively; qtgw11a, which favorable alleles was from the recurrent parent, explained 9.2% and 6.49% of phenotypic variation of TGW under two conditions respectively, and the dominant effects reached to 3.77g and 3.05g, which would be very important in hybrid breeding program; and qph5, which favorable alleles was from the donor parent, could explain 5.04% and 5.73% of phenotypic variation of PH under two conditions respectively. Besides, in this generation, three loci ie qpl3e, qpl4b and qpl4c were detected as well, which favorable alleles was from the donor parent, and could explain 7.87% and 11.02%, 11.76% and 14.25%, 12.43% and 12.28% of phenotypic variation of PL under two conditions respectively. |
PDF Full Text Request