The dissection of quantitative traits into their underlying Mendelian factors has become possible with the aid of molecular markers. In this study, we mapped and characterized quantitative trait loci (QTL) affecting testcross performance of maize (Zea mays L.) and discussed the consistency of these QTL across environments and testers. Two homozygous flint inbred lines were crossed to produce 380 F2 individuals which were genotyped at 89 restriction fragment length polymorphisin (RFLP) marker loci. By selfing the F2 plants 380 F3 lines were produced and subsequently crossed to two diverse dent inbred testers (T1 and T2). Each series of testcrosses (TC) was evaluated in field trials with two replications in four environments. Three traits were analyzed: protein content, kernel weight, and plant height. Genotypic and genotype-by-environment interaction variances were highly significant for all traits in both TC series and heritabilities ranged between 0.71 and 0.85. Genotypic correlations between the two TC series were 0.80 (protein content), 0.88 (kernel weight), and 0.84 (plant height). The method of interval mapping was used for characterization of QTL. Protein content was significantly affected by four QTL in TC with Tl explaining 32% of the phenotypic variance, (sigma(p)2) and five QTL in TC with T2 explaining 42% of sigma(p)2. For kernel weight, eight putative QTL were found in TC with T1 and six were found for T2, explaining about 60% of sigma(p)2. Seven QTL explained 52% and four 60% of sigma(p)2 for plant height. Results from QTL mapping agreed well across environments. Highly consistent results across testers were obtained for kernel weight and plant height but not for protein content. No significant evidence for epistasis among QTL was found.
引用
收藏
页码:378 / 389
页数:12
相关论文
共 34 条
[31]
Weir B.S., 1990, GENETIC DATA ANAL, V18, P639, DOI 10.2307/2533134