Genome-wide selection footprints and deleterious variations in young Asian allotetraploid rapeseed

Abstract

Brassica napus (AACC, 2n = 38) is an important oilseed crop grown worldwide. However, little is known about the population evolution of this species, the genomic difference between its major genetic groups, such as European and Asian rapeseed, and the impacts of historical large-scale introgression events on this young tetraploid. In this study, we reported the de novo assembly of the genome sequences of an Asian rapeseed (B. napus), Ningyou 7, and its four progenitors and compared these genomes with other available genomic data from diverse European and Asian cultivars. Our results showed that Asian rapeseed originally derived from European rapeseed but subsequently significantly diverged, with rapid genome differentiation after hybridization and intensive local selective breeding. The first historical introgression of B. rapa dramatically broadened the allelic pool but decreased the deleterious variations of Asian rapeseed. The second historical introgression of the double-low traits of European rapeseed (canola) has reshaped Asian rapeseed into two groups (double-low and double-high), accompanied by an increase in genetic load in the double-low group. This study demonstrates distinctive genomic footprints and deleterious SNP (single nucleotide polymorphism) variants for local adaptation by recent intra- and interspecies introgression events and provides novel insights for understanding the rapid genome evolution of a young allopolyploid crop.

Keywords: Asian rapeseed; allopolyploid; deleterious variations; introgression; selection footprints.

Figure 1

Pedigree and the genome assembly of NY7. (a) Asian rapeseed breeding history with the NY7 pedigree. Asian rapeseed experienced B. rapa introgression during the double‐high breeding stage (top panel) and then European rapeseed introgression during the double‐low breeding stage (low panel). In the low panel, only the three B. napus accessions (Tapidor, Darmor‐bzh and ZS11), which have been de novo sequenced, are shown. SL‐1: Shengliyoucai‐1; CDA: Chengduaiyoucai; CY2: Chuanyou2; NY1: Ningyou1; NY7: Ningyou 7; ZS11: Zhongshuang11. (b) Circos plot of the NY7 genome assembly with a genetic map, annotation and QTL. Circles inwards: pseudo‐chromosomes (black or white bars represent individual scaffolds), genetic maps, gene density, repeat density and QTL based on the Tapidor×NY7 population (Bna TNDH). (c) Genomic synteny and variations between the two de novo assembly genomes, NY7 and Darmor‐bzh.

Figure 2

Genetic divergence of Asian and European rapeseeds. (a) Phylogenetic tree and (b) principal component analysis of Asian (AS) and European (EU) rapeseeds. Asian rapeseeds were separated into two groups: double‐high (AS_DH) and double‐low (AS_DL). The de novo sequenced cultivars are indicated by dots.

Figure 3

Demographic inference for Asian rapeseeds. (a) The best demographic models for the origin of Asian (AS) and its double‐high (AS_DH) and double‐low (AS_DL) varieties. EU, Br and Bo refer to European rapeseed and its two progenitors, B. rapa and B. oleracea, respectively. (b) Divergent time (generation) and effective population size (Ne) changes for the European (EU) and Asian double‐high (AS_DH) rapeseed populations estimated using SMC++.

Figure 4

Genomic selection signals in Asian rapeseeds. (a) Significantly divergent genomic regions of Asian double‐high rapeseeds (AS_DH) relative to the European rapeseeds (EU) are shown on the top of the subfigure, and significant divergent genomic regions between Asian double‐low rapeseeds (AS_DL) and AS_DH are shown at the bottom. Selected genes and QTL found previously in those regions are indicated. (b) Local ancestry inference for Asian rapeseed. Left: Local ancestry inference for the 68 Asian rapeseeds in chromosome A10 (red: B. rapa genetic background; blue: EU genetic background). The red bars on the top indicate that the introgression events from B. rapa into the Asian rapeseed population have been almost fixed (i.e. found in >90% samples). Right: Local ancestry inference for the 30 Asian double‐low rapeseeds in chromosome A08 (blue: EU genetic background; yellow: Asian double‐high genetic background). The X‐axis represents the genomic locations of each chromosome. The Y‐axis presents representative individuals used for the ancestry inference. For details of all chromosomes, see Figure S14.

Figure 5

Identity‐by‐descent (IBD) inheritance pattern for the NY7 pedigree. For each chromosome, the genomic contributions by the four different parental lines to NY7 were colour‐coded as CY2 (yellow), CDA (orange), NY1 (green) and SL‐1 (purple). The accumulated percentages of genomic contribution by the four parental lines are shown in the pie graph. The genomic contribution of CY2 to NY7 was further divided into those from SL‐1 (36.38%) and CDA (9.79%).

Figure 6

Genetic load estimation for the European and Asian groups. The top panel shows the frequency distribution of deleterious SNPs (dSNPs) in the A subgenome of European (EU) and two Asian groups (AS_DH and AS_DL) with the arrow indicating the fixed dSNPs (with a frequency of 0.95‐1 in the target population). The bottom panel illustrates the relative frequency of deleterious to neutral variants (iSNPs) in the three groups.