The genetics of phytate content and morphological traits in Brassica rapa

In this thesis molecular genetic studies on Brassica rapa are described based on a collection of 256 accessions and 6 segregating populations. In chapter 2 and 3 the genetic variation and population structure are characterized in a set of genotypes from different geographical origins representing the various cultivar types. The association of traits with AFLP markers is analyzed in chapter 3, and the development of various mapping populations, linkage map construction and QTL analysis of phytate and diverse morphological traits are described in chapter 4 and 5.Genetic diversity in the primary gene pool of B. rapa was assessed by AFLP fingerprinting (chapter 2). The relationship between accessions was revealed by the dendrogram,indicating thatdifferent cultivar groups are often region(East Asia vs.Europe)specific, emerging independently in the different geographic regions. The UPGMA dendrogram (chapter 2) and population structure (chapter 3) showed that grouping based on markers coincided often with morphotypes (cultivar groups) within a geographic region. However this clustering was not supported by significant bootstrap values, suggesting that only a few genes are involved in causing the large differences in morphologies.A number of markers differed between subpopulations thereby providing an example oflinkage disequilibrium (LD) due to population structure.Diversity collections can be used to survey the allele frequencies and to explore associations with traits (chapter 3). The STRUCTURE program indicated an association between population structure and trait values even for traits such as phytate and phosphate levels.The population structureshould therefore be considered in association mapping studies in B. rapa .Markers that show association after correction for substructure can reliably be interpreted as being linked. Using the Composite Interval association mapping procedure, 54 markers were found significantly associated with at least one of the 9 different traits that were analyzed, 16 of which had known map positions. The observation that some of the marker-trait associations were confirmed by QTL analysis indicated that association mapping allows the detection of linkage with moderately frequent alleles.A multiple population approach was applied in QTL analysis for phytateandmorphologicaltraits, in which the populations were developed from wide crosses and alleles of multiple parental accessions can be evaluated,revealing a large number of genomic regions harboring allelic variation for traits of interest. For phytate and phosphate contents in seeds and leaves, and 20 morphological traits involved in flowering time, seed traits, plant height, leaf traits and turnip formation we identified a total of 54 QTL positions. Co-location of QTL for different traits was found in many cases, which might suggest pleiotropy or tight linkage. The analysis of co-location of QTL for phytate/phosphate and parental allelic effects suggested that these QTL control total phosphorus levels or are specific for the phytate pathway in B. rapa. Principal component analysis and co-localization of QTL for morphological traits indicated that some components of the genetic control of certain leaf and seed-related traits, flowering time and turnip formation might be based on the same genes, indicating a commongenetic regulation of correlated traits underlying these QTL.The comparative analyses of QTL for certain traits, such as phytate and phosphate contents and flowering time, between the 6 populations and with Arabidopsis thaliana or other related Brassica species are discussed.