| Comparing with the other important component factors in maize yield, kernel row number(KRN), has the higher broad heritability, while less bias in phenotyping. These kinds of characteristics offer KRN critical advantages in QTL mapping and genetic basis dissection. In previous studies, we identified five major QTL using the F2:3 populations developed by the four-rowed waxy corn(SLN) and Nong531(N531). In this project, we deepen mapped the five major QTL by constructing a series of backcross populations from the same cross. We also analyzed the interactions among the QTL, proposed a regulation network in kernel row number. Finally, we suggested three candidate genes for one of the most convincing QTL qKRN5.04. The main results were as follows:1. Five major QTL for KRN expressing in different environments were selected respectively based on the mapping result of F2:3 from the cross of four-rowed waxy corn and Nong531. The effects were validated in BC3F2 and BC4F2 population with the SLN as the donor parent and Nong531 as the recurrent parent. Then the different interval was increased markers to narrow down and the qKRN2.04 was narrowed from 15 Mb to about 5.3Mb, qKRN4.09 from 17 Mb to 1.3Mb, qKRN5.04 from 69 Mb to 3.7Mb, qKRN8.05 from 3Mb down to 1.0Mb, qKRN9.03 reduced from 64 Mb to 17 Mb within range, which laid the foundation for further fine mapping. What important is that the qKRN5.04 was almost consistently under two environments(two adjacent interval position umc1060-N5M2 and N5M2-N5M5), and explain the phenotypic variation 6.44%-21.76%. Therefore qKRN5.04 was determined the key segment for fine mapping and Map-based cloning.2. To know the interaction among the five major QTL, we analyzed a 504 BC3F3 population combined with their pedigree were used to conduct the pairwise analysis of variance for multiple comparisons for the five main QTL. The results showed that the QTL-QTL interactions of qKRN2.04-qKRN4.09, qKRN2.04-qKRN9.03, qKRN4.09-qKRN5.04, qKRN4.09-qKRN9.03, qKRN8.05-qKRN9.03 rendered significant differences in three environments. At the same time, the plant and pedigree results were used for the five different main effect loci of two, three, four and five combinations. It can be found that only the pairwise interaction of qKRN2.04-qKRN5.04 can be detected at three stable environment; there are two groups three loci interaction combinations, qKRN2.04-qKRN5.04-qKRN8.05 and qKRN2.04-qKRN5.04-qKRN9.03 in three environments; there were a set of interactions four loci combination, qKRN2.04-qKRN5.04-qKRN8.05-qKRN9.03 has been stably detected at three environment; another two groups four loci interaction combination qKRN2.04-qKRN4.09-qKRN5.04-qKRN8.05 and qKRN2.04-qKRN4.09-qKRN8.05-qKRN9.03 were stable under two environments. What was expected was the five loci interaction combination qKRN2.04-qKRN4.09-qKRN5.04-qKRN8.05-qKRN9.03 has been stable at three environments. Considering the effects analysis for different major effect locus interactions, we propose that the kernel row number was affected by a complex interaction network mode regulation.3. QTL for KRN was identified by the inclusive complete interval mapping(ICIM) with single environment and joint-environment. The results showed that the major effect QTL for qKRN5.04 was located to 136.3-140.0Mb through the advanced backcross populations, explaining the biggest phenotypic variation up to 21.76% and the effect value of 0.80-1.76 row. While the effect of joint-environment for QTL mapping can explain the phenotypic variation up to 24.01%. Finally, the qKRN5.04 was remapped within the region of about 300 kb by the analysis of different recombinants. By bioinformatics analysis, we proposed that there are three candidate genes, GRMZM2G300841, GRMZM5G897364 and GRMZM2G060253. Based on the published RNA expression data, all of these three candidates showed specific expression pattern in maize tissues, which indicates their potential function in the regulation of kernel row number. |
PDF Full Text Request