High-Density Genetic Linkage Map Construction And QTLs Analyses For Phenotypic Traits In Tree Peony

Belonging to Paeonia of the Paeoniaceae, tree peonies (Paeonia Sect. Moutan) are Chinese traditional famous flowers, and its breeding is important basic for the sustainable development of tree peony industry. Therefore, understanding of the molecular mechanisms underlying allelic variation of phenotypic character is considered as one of the main prerequisites for tree peony genetic breeding improvement, transformed from the traditional hybrid breeding to the modern molecular breeding. In this study, we selected an optimizing segregation population, derived from the cross P. ostti ’Feng Dan Bai’ M24 x P.× suffruticosa ’Hong Qiao’, for genetic linkage mapping, and constructed the first high-density genetic linkage map of tree peonies using genotyping by specific-locus amplified fragment sequencing (SLAF-seq) and simple sequence repeat (SSR) markers. On this bases, the quantitative trait loci (QTLs) for main traits of tree peony were also detected. The main results of this dissertation are indicated as follows:(1) The levels of polymorphisms between parents of three segregation populations were detected by employing 19 pairs of SSR primers. The results showed that,19 pairs of SSR primers could produce most informative polymorphism loci of 27 between the ’Feng Dan Bai’ M24 and ’Hong Qiao’ among the parents of three segregation populations. The genetic distance between ’Feng Dan Bai’ M24 and ’Hong Qiao’ was 0.7070, which was also farthest among the parents of three segregation populations. Basis on it, segregations of genotype in 195 progeny of selecting mapping population were investigated, using SSR markers.73.33% of genotype in mapping population segregated in an expected Mendelian ratio (P0.05). Therefore, the segregation population of ’Feng Dan Bai’ M24 × ’Hong Qiao’ was selected for the construction of genetic linkage map in tree peony.(2) Illumina Solexa sequencing of reduced-representation libraries generated a total of 285,403,225 reads, about 78 Gb raw data, from parents and 195 Fi progenies. Reads were aggregated into 309,198 SLAF tags, of which 85,124 (27.5%) were polymorphic. After genotype calling and removal of tags with incomplete data or unsuitable for CP (cross-pollinator) population, a total of 3518 SLAF markers were finally identified as effective markers for use in constructing a linkage genetic map. Chi-square test showed that 1854 markers were distorted from Mendelian segregation ratios (P<0.05), accounted for 52.70%.(3) A set of 400 SSR primers developed from different DNA resources of tree peonies, were screened in the two parents and six Fi individuals. Using 1% agarose gel electrophoresis,365 pairs of SSR primers, effective PCR, were detected, counted about 91.2%. Using 6% polyacrylamide gel electrophoresis,79 pairs of SSR primers, polymorphisms between parents, were detected, counted about 21.64%. These polymorphisms markers were fluorescent-modified, and genotyped in tree peonies mapping population. The results indicated that all 79 markers divided into five segregation types. A chi-square test was performed on marker genotyping data for mapping population, resulting in that 26 markers were distorted from Mendelian segregation ratios (P<0.05), accounted for 32.91%.(4) A total of 3597 demonstrably heterozygous markers, including 3518 SLAFs and 79 SSR, were available for linkage analysis, and the first high-density genetic linkage map of tree peonies were constructed. This genetic linkage map comprising 1261 markers distributed over five linkage groups, included 1189 SLAFs and 72 SSR. The map spanned 1061.94 cM with an average inter-marker distance of 0.84 cM. The genetic length of LGs ranged from 108.947 cM (LG3) to 296.431 cM (LG2), with an average of 184.140 cM.(5) The statistical analysis for 27 phenotypic traits of mapping population, including stem, leaf, flower and follicle traits, indicated that all traits between the parents were significant differences, the genetic variation coefficient ranged from 10.93% to 78.56%. The mapping population displayed continuous distribution of phenotypes and normal segregation. A total of 49 QTLs, controlled 20 phenotype characters, were identified by restricted multiple-QTL model (MQM) method of MapQTL software. Proportion of phenotypic variance explain (PVE) for these QTLs was from 8.3% to 71.9%. Among these QTLs, QTL-pn-2 for petals number had the highest proportion of PVE, with 71.9%. Furthermore, using interval mapping, QTLs mapping of flower color were carried out. Three putative QTLs were detected for flower color, with a small proportion of PVE, from 1.27% to 30.02%.In conclusion, the results of this study will serve as an important foundation for genomic structure and function, map-based gene cloning, and marker-assisted selection breeding, and will further provide experiences for the analysis of genetic mechanism of important horticultural traits, and will facilitate the development of breeding and industry in tree peony.