Multi-omics analysis dissects the genetic architecture of seed coat content in Brassica napus

Abstract

Background: Brassica napus is an important vegetable oil source worldwide. Seed coat content is a complex quantitative trait that negatively correlates with the seed oil content in B. napus.

Results: Here we provide insights into the genetic basis of natural variation of seed coat content by transcriptome-wide association studies (TWAS) and genome-wide association studies (GWAS) using 382 B. napus accessions. By population transcriptomic analysis, we identify more than 700 genes and four gene modules that are significantly associated with seed coat content. We also characterize three reliable quantitative trait loci (QTLs) controlling seed coat content by GWAS. Combining TWAS and correlation networks of seed coat content-related gene modules, we find that BnaC07.CCR-LIKE (CCRL) and BnaTT8s play key roles in the determination of the trait by modulating lignin biosynthesis. By expression GWAS analysis, we identify a regulatory hotspot on chromosome A09, which is involved in controlling seed coat content through BnaC07.CCRL and BnaTT8s. We then predict the downstream genes regulated by BnaTT8s using multi-omics datasets. We further experimentally validate that BnaCCRL and BnaTT8 positively regulate seed coat content and lignin content. BnaCCRL represents a novel identified gene involved in seed coat development. Furthermore, we also predict the key genes regulating carbon allocation between phenylpropane compounds and oil during seed development in B. napus.

Conclusions: This study helps us to better understand the complex machinery of seed coat development and provides a genetic resource for genetic improvement of seed coat content in B. napus breeding.

Keywords: Brassica napus; Co-expression network; Phenylpropane pathway; Seed coat content; Seed oil content; TWAS; eQTL.

Fig. 1

Transcriptome-wide association studies of SCC. a, b Manhattan plot of TWAS results (FDR < 0.05) for SCC at 20 DAF (a) and 40 DAF (b). Each point represents a single gene tested. Genomic positions of genes are plotted on the X-axis and the log-transformed FDR values of association between gene expression and SCC are plotted on the Y-axis. The genes positively or negatively associated with the SCC are plotted above or under the black bold line, respectively. The golden dots indicate genes involved in the phenylpropanoid biosynthesis pathway and secondary cell wall development identified in Arabidopsis. The dashed gray horizontal lines represent the significance level. c Venn diagram showing shared genes between significant genes identified by TWAS at 20 DAF and 40 DAF. d GO enrichment analysis of overlapped significant genes identified by TWAS at 20 DAF and 40 DAF. The dot size and color indicate the gene number and the range of FDR values, respectively

Fig. 2

SCC-related gene modules and co-expression networks in seeds. a Gene correlation network of SCC-related module (M65) at 20 DAF. Phenylpropanoid biosynthesis-related genes significantly associated to SCC detected by TWAS at 20 DAF are labeled in the module. b Gene correlation network of SCC-related modules (M97, M79, M139) at 40 DAF. Phenylpropanoid biosynthesis-related genes significantly associated to SCC detected by TWAS at 40 DAF are labeled in the module. c Correlation between the expression pattern (X-axis) of M65 and SCC (Y-axis). d Correlation between the expression pattern (X-axis) of M139 and SCC (Y-axis). e Correlation network built on the overlapped genes of the TWAS-significant genes, M65 and M139. The green dots represent phenylpropanoid biosynthesis-related transcription factors and red dots represent other phenylpropanoid biosynthesis-related genes, the blue dots represent the rest of genes. The size of a node is proportional to its degree (the number of correlated genes). Genes in a, b, and e are shown by dots and gene-gene correlations (FDR < 0.01) are indicated by gray lines

Fig. 3

The regulatory hotspot on chromosome A09 affects SCC. a Manhattan plot of GWAS results for SCC. Three significant QTLs are denoted. b Manhattan plot of GWAS results for expression value of BnaA09.TT8, BnaC09.TT8 and BnaC07.CCRL. The SCC QTLs qSCC.A09 identified by GWAS are denoted. Significant eGWAS associations for TWAS-significant genes at 40 DAF are marked (gray lines); associations co-localized with qSCC.A09 are highlighted in green. BnaA09.TT8, BnaC09.TT8 and BnaC07.CCRL localization results are highlighted in black. c The log-transformed P values form GWAS plotted against those for variants from eGWAS of qTT8.A09 (around qSCC.A09 lead SNP 150 kb). d Flowchart of screening of putative phenylpropanoid and lignin synthesis-related genes regulated by BnaTT8s. Based on the eGWAS results of phenylpropanoid and lignin synthesis-related genes, whose eQTLs co-localized with qSCC.A09 were detected. eQTLs of some genes were not detected to be co-localized with qSCC.A09 using BnaTT8s as a covariate. Among these genes, differentially expressed genes when BnaTT8s were mutated were identified

Fig. 4

Functional characterization of BnaTT8 as a positive regulator of seed lignin biosynthesis in B. napus. a Seeds of CRISPR/Cas9-induced BnaA09.TT8 and BnaC09.TT8 homozygous double mutant lines (L14, L17, L21, L18, L24, and L30). Bar = 1 cm. b the seed coats of L21 and L24. Bar = 1 cm. Determination of thickness of seed coat (c), SCC (d), lignin content (e), and seed oil content (f) in the BnaTT8 double knockout mutant seeds and the control (WT). Values are means ± SD of three biological replicates (n = 6). Student’s t-test was used for statistical analysis between the BnaTT8-sgRNA lines and WT (*, P < 0.05)

Fig. 5

Functional characterization of BnaCCRL as a positive regulator of seed lignin biosynthesis in B. napus. a Seeds of CRISPR/Cas9-induced BnaC07.CCRL and BnaA03.CCRL homozygous double mutant lines (L35, L43, L48) and BnaC07.CCRL, BnaA03.CCRL and BnaC05.CCRL homozygous triple mutant line L21. Bar = 1 cm. b the seed coats of L43 and L21. Bar = 1 cm. Determination of thickness of seed coat (c), SCC (d), lignin content (e), and seed oil content (f) in the BnaCCRL double or triple mutant seeds and the control (WT). Values are means ± SD of three biological replicates (n = 6). Student’s t-test was used for statistical analysis between the BnaCCRL lines and WT (*, P < 0.05)

Fig. 6

Correlation analysis between TWAS significant gene expression and SOC trait or SCC trait. a Venn diagram showing shared genes between significant genes identified by TWAS for SOC and SCC. b Correlation analysis between TWAS significant gene expression and SOC or SCC at 40 DAF. Each dot represents a gene. c Regulatory network of key genes for carbon source allocation in B. napus