Genetic and phenotypic analysis of complex seed and root traits in oilseed rape (Brassica napus L.)

Abstract

Rapeseed (Brassica napus L.) is an important oilseed crop. Its oil is used for human consumption, as green fuel (biodiesel), and in the chemical and pharmaceutical industry. The cake and meal, residues of oil pressing and extraction, are used as valuable components for feeding animals. Seed metabolism and root traits are two important components of seed quality and yield, respectively, in B. napus. Both are controlled by complex genetic mechanisms. The aim of this study was to develop and use novel high throughput DNA sequencing techniques, on one hand, to investigate digital gene expression (DGE) of gene networks responsible for important seed quality traits, and on the other hand to use DGE along with high-throughput phenotyping of root architectural traits in a rapeseed population for marker trait association. Two B. napus black seeded winter genotypes were selected to investigate their transcriptome during seed development by DGE: Express617 which is a 00-quality cultivar with high lignin content and V8, which has high seed erucic acid and glucosinolate (GSL) contents but moderate levels of seed lignin, another antinutritive compound in the seed meal. Early stages of seed development from 2-28 dap showed high correlation and a great variation was observed in later stages and the seed maturation phase in the two genotypes. Based on their previously known differential phenotypes for total GSL and lignin, differential expression of genes involved in GSL metabolic and phenylpropanoid/lignin biosynthesis pathway was observed in the MapMan metabolic pathway annotator during seven seed developmental time points (2-84 days after pollination). A total of 58 genes were annotated to the phenylpropanoid pathway specifically directed to lignin biosynthesis, while 34 genes were detected from the GSL metabolic pathway. Expression pattern of genes involved in lignin biosynthesis showed up-regulation in Express617 during the seed maturation phase from 42 until 70 dap. Expression of six MYB family transcription factors (MYB4, 32, 58, 61, 63 and 85) were detected which are responsible for activation of monolignol pathway genes by coordinating binding motifs corresponding to their AC-rich elements. AC elements are present in the promoters of most monolignol pathway genes, including PAL, 4CL, C3H, CCoAOMT, CCR and CAD, and many of these genes showed different expression patterns between the two genotypes corresponding to their differences in lignin compounds. Similarly, several genes known to be involved in the GSL metabolic pathway are considered as candidates for controlling GSL contents in rape seeds to reduce total glucosinolates in rapeseed meal. In this study, expression of 34 genes was observed were involved in GSL amino acid side chain elongation, core structure formation from amino acid moiety, secondary modification and degradation enzymes. Expression of transcription factors OBP2/DOF1.1 and the MYB family (MYB34, 28 and 51) which are known to be responsible for GSL pathway regulation were also detected during seed developmental stages. Further characterization of these genes in developing seeds from each time point and correlation with seed metabolomic profiles for GSL and lignin content will help in validation of their potential role in controlling seed fibre content in black seeded rapeseed genotypes. In this project, a digital root phenotyping system based on mini-rhizotrons was also optimized for phenotyping root traits in rapeseed. This non-destructive gel based system has facilitated the visualization of the root system in large B. napus population. Root architectural traits were studied in two sets of B. napus genotypes, Express617 x V8 DH population (94 lines) and 500 inbred lines in the diversity set include winter, spring, vegetable and swede types. Five root traits were investigated for phenotypic analysis, including primary root length (PRL), rate of primary root growth (RoG), lateral root length (LRL), lateral root number (LRN) and lateral root density (LRD). A large variation, segregation and heritability of root architectural traits in the bi-parental population proved that these are quantitatively traits controlled by multiple genes, which give intimation to proceed for genetic improvement and selection of lines with improved root system. In the bi-parental population Express617 x V8 DH, 11 QTL regions associated with root architectural traits were detected. Some of these are also co-localized with previously detected QTL associated for seedling and yield traits in the same population, and potentially also co-localized with biomass and yield-related traits in other populations. QTL mapping is a first step for marker assisted selection, as well as for map-based gene discovery. In a large B. napus diversity set, 38 significant marker-trait associations were detected for root architectural traits. LD analysis confirms the low overall level of linkage disequilibrium in B. napus which could be due to controlled crosses followed by several generations of selfing to develop new varieties with improved seed qualities. With the availability of the B. napus genome sequence, knowledge of sequence variation in specific genes and cost-effective high-throughput genotyping and phenotyping of qualitative and quantitative traits, it is expected that molecular breeding will play an important role in future breeding of new cultivars.