Mapping traits of agronomic interest in kiwifruit and development of markers for Markers assisted selection (MAS)

In kiwifruit, dioecy and intraspecific polyploidy are great challenges in breeding and cultivar improvement. Marker-assisted selection (MAS) can enhance the selection efficiency and precision in kiwifruit breeding. The prerequisite to MAS is the construction of a high-density linkage map allowing linkage association between marker loci and quantitative trait loci (QTLs) of particular traits or genes of interest. Consequently, the production of dense maps in kiwifruit and mapping of traits, in particular, gender determinant and fruit quality traits was the aim of this thesis. Thus, a mapping population was generated from a cross between ‘C8’ (female) and ‘A54.19’ (male), both diploid kiwifruit (A. chinensis), genetically and geographically unrelated. The work of thesis is composed of five chapters. In the first chapter, a literature review that supports the main aspects in kiwifruit breeding is presented. In the second chapter, we report the construction of genetic maps by merging ddRAD SNPs and SSRs markers. Results showed that even in the presence of a reference genome (Huang et al. 2013), our genetic maps help to improve the genome assembly. Furthermore, the saturated maps help to restrict the chromosomal region where genes are located and to find genetic loci that control Mendelian traits like the gender. Thus, we have isolated four SSR markers by screening the contigs of sex-determining region. These markers were able to discriminate between male and female progeny, but they cannot be used to screen germplasm collection where the parentage of accessions is not known in advance. In the third chapter, quantitative trait loci (QTLs) analysis traits related to fruit quality were carried out using fruit from mapping population. Many QTLs had large effects in terms of explained variance, however, they showed poor stability probably due to the limited size of the progeny. In the fourth chapter, estimation of genetic variance and selection of superior male parents for their capability of transferring traits of interest to progeny were performed using best linear unbiased prediction (BLUP) and REML in a controlled mating design, involving 35 small cross families. This analysis include the estimation of genetic component of variance, heritability, the combining ability of cross parents and breeding value for several vegetative and fruit quality traits in kiwifruit. In the last chapter, we describe the sequencing of male parent of our mapping population and the future steps towards the exploration of the sex-determining region