| Rapeseed is one of the most effective oil crops for oil production. China is the largest rapeseed producer in the world, in which more than 80% of the plant area is occupied by Brassica napus. In the middle and lower valley of the Yangtze River in South China, the largest cultivated region for rapeseed, the soil available P concentration is usually lower than 10 mg/kg. However, B. napus has high P requirement for its optimal seed yield and quality. Hence, the development of cultivars with enhanced phosphorus (P) use efficiency would be an economical and sustainable way for the management of P nutrition in B. napus production. In our previous research, a set of 124 F10 B. napus recombinant inbred line (RIL) population named the BE-RIL population was developed using the single-seed descent method from a cross between P-inefficient cultivar'B104-2 (B)'and P-efficient cultivar'Eyou Changjia (E)'.In this study, an improved BE-RIL genetic linkage map was constructed using gene-based marker (GBM), sequence-related amplified polymorphism (SRAP) and simple sequence repeat (SSR). Two-year field trials as well as two hydroponic trials were conducted with normal P and low P treatments. Phenotypic variation of seed yield and yield-related traits and seed mineral concentrations (P, Ca, Mg, Zn, Cu, Fe, Mn) at maturing stage, dry weight and P concentration at seedling stage were investigated. Quantitative trait loci (QTL) analysis was performed with the purpose to unravel the genetic network of low-P response of B. napus at seedling and maturing stages.A total of 243 SSR markers were developed from 171 bacterial artificial chromosome (BAC) end sequences and/or B. rapa seed BAC sequences.256 GBMs was developed from 46 functional genes involved in Arabidopsis thaliana P homeostasis. In total,49 GBMs corresponding to 26 genes, assigned to eight functional categories, were integrated into the BE-RIL map. Genotyping data generated in this study were attempted for integrating the marker loci into the framework linkage map using JoinMap software Version 4.0. The present map has a total of 840 loci, including 62 amplified fragment length polymorphism (AFLP),257 SRAP,472 SSR and 49 GBMs. The total map length was 1913.6 cM with an average distance of 2.3 cM between two loci. Based on the identified 24 conserved chromosomal blocks on A. thaliana genome, a total of 115 conserved blocks, including 38 syntenic blocks and 77 insertion segments, were identified using 304 of the 840 linked markers with known sequence information. According to physical positions on the Arabidopsis genome,3162 orthologs of 789 Arabidopsis functional genes related with target traits were mapped onto syntenic blocks and insertion segments. This high-density integrated molecular genetic map will be useful for identifying QTL that control P homeostasis and putative candidate genes for the efficient use of P in B. napus.Compared with P-inefficient cultivar'B104-2', P-efficient cultivar'Eyou Changjia' showed a higher value for dry weight and P concentration under low P condition at seedling stage as well as a higher value for seed yield and yield-related traits at maturing stage, except seed weight. In addition, a higher accumulation of minerals was observed in seeds of P-efficient cultivar'Eyou Changjia'than P-inefficient cultivar'B104-2'. However, for all the tested traits, there was no significant difference under normal P condition. The frequency distributions of all the traits showed continuous phenotypic variation for the BE-RIL population, and significant transgressive segregation was observed in both directions, suggesting multiple gene action. Correlation coefficients among traits under low P and normal P condition were different, and a low-to-high broad-sense heritability for each trait was observed in the BE-RIL population. This indicated the genetic loci controlling these traits might be different under two P levels, and environmental influence for each trait might be different.Phenotype data for seed yield, seed weight, seed number, pod number, plant height, branch number and phosphorus efficiency coefficient (PEC) from two field trials was used for QTL detection by WinQTL cartographer 2.5 software as well as the mean values of the two crop seasons. A total of 95 putative QTL distributed across 16 chromosomes were identified, including 6 QTL for PEC,48 and 41 QTL under normal and low P conditions for the other six traits, respectively. These QTL accounted for a range of 7.3-25.4% of the total phenotypic variation.19 (20%) were detected in two seasons and in the mean value of two season. Based on the common molecular markers on different genetic maps,22 reported QTL, which were related to the same traits and identified from other genetic populations of B. napus, were projected onto the BE-RIL genetic map using the map projection function of BioMercator 2.1 software. These QTL co-located with 18 QTL for corresponding traits detected in the present study, indicating a potential QTL hotspot for each trait in different genetic backgrounds and environmentsA total of 78 putative QTL, including 45 under normal P conditions and 33 under low P conditions, were detected in the two-year field trials for seven mineral traits. These QTL were mainly located on 17 linkage groups, explaining 7.84%to 18.1% of the phenotypic variation. About 65% of these QTL were co-located with at least one other QTL.16 QTL were identified in at least two environments, but the rest were detected in only one environment.109 putative QTL were identified for shoot dry weight, root dry weight, P concentration at seedling stage, including 49 under normal P conditions and 60 under low P conditions. These QTL scattered across 17 chromosomes except A10 and C2, and accounted for a range of 6.9-26.7% of the total phenotypic variation. Only one QTL on A6 was observed in three trials under high P level, suggesting culture environment and growth media might show heavy influence on QTL expression for the same trait. Some QTL also expressed at both seedling and maturing stages, suggesting the same gene action during plant development.Analysis by synthesis of 282 putative QTL, we found QTL were clustered on the location of each linkage group, and co-located with QTL detected previously in our lab, such as on chrosome A0, A3, A6, C3 etc. This indicated QTL hotspots existed on different linkage groups controlling different traits. By comparative mapping between Arabidopsis and B. napus,224 orthologs of 198 genes involved in Arabidopsis P homeostasis and/or controlling yield-related traits were associated with 50 QTL, and 21 genes involved in ion homeostasis in Arabidopsis were mapped to 13 QTL intervals. In addition,14 GBMs developed from genes involved in Arabidopsis P homeostasis were mapped to target QTL intervals, of which five were located in the confidence intervals of QTL detected under low P condition. These results may provide useful information for improving P efficiency of B. napus in soils with low P availability. |
PDF Full Text Request