Precise editing of CLAVATA genes in Brassica napus L. regulates multilocular silique development

Abstract

Multilocular silique is a desirable agricultural trait with great potential for the development of high-yield varieties of Brassica. To date, no spontaneous or induced multilocular mutants have been reported in Brassica napus, which likely reflects its allotetraploid nature and the extremely low probability of the simultaneous random mutagenesis of multiple gene copies with functional redundancy. Here, we present evidence for the efficient knockout of rapeseed homologues of CLAVATA3 (CLV3) for a secreted peptide and its related receptors CLV1 and CLV2 in the CLV signalling pathway using the CRISPR/Cas9 system and achieved stable transmission of the mutations across three generations. Each BnCLV gene has two copies located in two subgenomes. The multilocular phenotype can be recovered only in knockout mutations of both copies of each BnCLV gene, illustrating that the simultaneous alteration of multiple gene copies by CRISPR/Cas9 mutagenesis has great potential in generating agronomically important mutations in rapeseed. The mutagenesis efficiency varied widely from 0% to 48.65% in T₀ with different single-guide RNAs (sgRNAs), indicating that the appropriate selection of the sgRNA is important for effectively generating indels in rapeseed. The double mutation of BnCLV3 produced more leaves and multilocular siliques with a significantly higher number of seeds per silique and a higher seed weight than the wild-type and single mutant plants, potentially contributing to increased seed production. We also assessed the efficiency of the horizontal transfer of Cas9/gRNA cassettes by pollination. Our findings reveal the potential for plant breeding strategies to improve yield traits in currently cultivated rapeseed varieties.

Keywords: Brassica napus; CLAVATA genes; CRISPR/Cas9; genome editing; multilocular silique.

Figure 1

BnCLV gene models with target sequences and schematics of binary plasmid vectors. (a) The BnCLV3 gene model includes three exons (white box) separated by two introns (represented by the solid line). The vertical line in the gene model indicates the target site, and the arrow indicates the sgRNA direction. The target sequences are shown with the PAM highlighted in red. (b) The constructs of SCLV3 and UCLV3 house the following: a hygromycin resistance cassette consisting of the hygromycin phosphotransferase coding sequence driven by the cauliflower mosaic virus 35S promoter; a Cas9 expression cassette comprising the sequence encoding Cas9 driven by P_35S or a ubiquitin promoter from maize; and two sgRNAs S1 and S2 driven by the U3b and U6‐1 promoters from Arabidopsis, respectively. (c, b) The BnCLV1 gene model with target sites S3 to S6 and the BnCLV2 gene model with target sites S7 to S10. (e) The binary constructs SCLV1 and SCLV2 with four sgRNAs driven by the U3b, U3d, U6‐1 and U6‐29 promoters from Arabidopsis. [Colour figure can be viewed at wileyonlinelibrary.com]

Figure 2

Detection of mutations in BnCLVs using PAGE method. (a) The transgenic positive detection in T₀ regenerated plants via a PCR assay using NPT II gene‐specific primers and J9707 (WT) as a negative control. (b) Detection of the targeted mutations in T₀ plants with the WT as a negative control. (c) Detection of mutations in different targets in the two BnCLV1 copies in T₀ plants with the WT as a negative control. (d) Detection of mutations in different targets in the two BnCLV2 copies in T₀ plants with the WT as a negative control. The numbers ‘S#‐#’ and ‘U#‐#’ above the PAGE gels represents the corresponding individual IDs ‘SCLV#‐#’ and ‘UCLV#‐#’, respectively. ‘S#’ represents the specific targets of BnCLVs. A red arrow indicates that the tested target has been edited. [Colour figure can be viewed at wileyonlinelibrary.com]

Figure 3

Detailed genotype analysis of mutations in BnCLV3 on S1 in T₀ generation. (a) Sequencing results of 22 editing T₀ plants at the S1 site. Insertions and deletions are indicated in red font and with red hyphens, respectively. Edited plants with red stars have multilocular phenotypes. On the left, A and C and the WT allele of the BnA04. CLV3 and Bn0C4. CLV3 copies, respectively; a# and c# show the mutant allele numbers. ‘−’ and ‘+’ indicate the deletion and insertion of the indicated number of nucleotides, respectively; ‘−/+’ indicates the simultaneous deletion and insertion of the indicated number of nucleotides. (b) Mutation types and frequency at the S1 target site in 22 T₀ plants. In the left insert table, the occurrence of deletions (d), insertions (i) and combined (c) mutation types is shown. In the right insert table, the frequency of different mutation lengths is shown. In the middle insert table, frequency of different insertion types is shown. X‐axis: d#, # of base pair (bp) deleted from the target site; i#, # of bp inserted at the target site; c#, combined mutation. [Colour figure can be viewed at wileyonlinelibrary.com]

Figure 4

Germ‐line transmission of CRISPR/Cas9‐induced mutations at the S1 target site of SCLV3 from the T₀ generation to the T₂ generation. CRISPR/Cas9‐induced insertions and deletions are indicated by red font and red hyphens, respectively. On the left, A and C and the WT allele of the BnA4. CLV3 and BnC4. CLV3 copies, respectively; a# and c# show the mutant allele numbers. ‘−’ and ‘+’ indicate the deletion and insertion of the indicated number of nucleotides, respectively. [Colour figure can be viewed at wileyonlinelibrary.com]

Figure 5

Phenotypes of the BnCLV3 mutants. (a) Leaf numbers of 30‐day‐old seedlings in WT, single and double homozygous mutants of BnCLV3. (b, c) The inflorescences (b) and SAM (c) in the WT and a double homozygous mutant of BnCLV3. (d) Cross sections of gynoecia in the WT and a double homozygous mutant of BnCLV3 at stages 9–10. (e) Siliques in the WT, single and double homozygous mutants of BnCLV3. Bar = 1 cm. (f) Statistical analysis of the leaf number, carpel number, silique length, silique thickness, NSS, thousand seeds weight and SW per silique in the WT and single and double homozygous mutants of BnCLV3. The data and error bars represent the mean ± SD (n ≥ 15 plants for each genotype). Upper‐case letters indicate a significant difference at the 0.01 probability level. aa, homozygous mutation of BnA04. CLV3; cc, homozygous mutation of BnC04. CLV3; aacc, double homozygous mutation of BnA04. CLV3 and BnC04. CLV3. [Colour figure can be viewed at wileyonlinelibrary.com]

Figure 6

Genome editing of cultivars by crossing Cas9‐sgRNA lines. (a) Selection of the natural outcrossing plants from open‐pollination progeny of HY. Plants with serrated leaves are hybrids of HY and the double homozygous mutant SCLV3‐35 for the incompletely dominant lobed‐leaf trait. (b) The read numbers at the S1 site of BnCLV3 in mixed genomic DNA from F₁ hybrid plants with T‐DNA are shown, and WT (HY) was included as a control. (c) Frequency of different genotype reads. Original mutant reads, reads with the same mutations detected in SCLV3‐35; novel mutant, reads with different mutations detected in SCLV3‐35.