CRISPR/Cas9-Mediated Targeted Mutagenesis of BnaCOL9 Advances the Flowering Time of Brassica napus L.

Abstract

Rapeseed (Brassica napus L.) is one of the most important oil crops in the world. The planting area and output of rapeseed are affected by the flowering time, which is a critical agronomic feature. COL9 controls growth and development in many different plant species as a member of the zinc finger transcription factor family. However, BnaCOL9 in rapeseed has not been documented. The aim of this study was to apply CRISPR/Cas9 technology to create an early-flowering germplasm resource to provide useful material for improving the early-maturing breeding of rapeseed. We identified four COL9 homologs in rapeseed that were distributed on chromosomes A05, C05, A03, and C03. We successfully created quadruple BnaCOL9 mutations in rapeseed using the CRISPR/Cas9 platform. The quadruple mutants of BnaCOL9 flowered earlier than the wild-type. On the other hand, the flowering time of the BnaCOL9 overexpression lines was delayed. An analysis of the expression patterns revealed that these genes were substantially expressed in the leaves and flowers. A subcellular localization experiment demonstrated that BnaCOL9 was in the nucleus. Furthermore, we discovered that two key flowering-related genes, BnaCO and BnaFT, were highly elevated in the BnaCOL9 mutants, but dramatically downregulated in the BnaCOL9 overexpression lines. Our findings demonstrate that BnaCOL9 is a significant flowering inhibitor in rapeseed and may be employed as a crucial gene for early-maturing breeding.

1. Introduction

Rapeseed is an important oil and economic crop worldwide. B. napus (AACC, 2n = 38) was generated 7500 years ago from a natural hybrid of Brassica rapa (AA, 2n = 20) and Brassica oleracea (CC, 2n = 18) and subsequently underwent chromosomal doubling to form an allotetraploid plant [1]. Rapeseed is classified into three ecotypes based on the environment required for flowering: winter, semi-winter, and spring. Depending on local circumstances, specific ecotypes can be planted in various regions. Therefore, the optimal flowering time for rapeseed planting is critical as it influences not only the production but also other crop rotations [2]. As a result, the flowering time is an essential feature for breeders throughout the rapeseed breeding process.

The flowering regulation network in Arabidopsis has been extensively researched; it primarily contains the photoperiod, vernalization, temperature, autonomy, age, gibberellin (GA), and sugar metabolism pathways [3,4]. Among these pathways, the flowering pathway contains several important genes known as integrators; for example, SOC1 (SUPPRESSOR OF OVEREXPRESSION OF CONSTANS1), CO (CONSTANS), and FT (FLOWERING LOCUS T). SOC1 encodes a MADS-box transcription factor that integrates flowering signals from the photoperiod, temperature, hormone, and age pathways [5]. CO was the first gene identified in the BBX family and regulates the flowering time via the photoperiod pathway [6]. FT encodes a phosphatidylethanolamine-binding protein, which is a small mobile protein. This gene is a major integrator in the control of the flowering pathway [7].

COL9 (CONSTANS-like 9) and CO belong to the BBX gene family and they encode zinc finger transcription factors [8,9]. There are 32 known B-BOX genes in Arabidopsis, which may be classified into five groups (I–V) based on amino acid variations in the B-BOX motif and CCT domain [9]. COL9 can suppress the expression of FT and CO, delaying the flowering time in Arabidopsis [10]. Other B-BOX genes that have been linked to flowering time regulation in Arabidopsis include BBX4/COL3, BBX6, BBX10, BBX19, BBX24, BBX30, BBX31, and BBX32 [11,12,13,14,15]. BBX32 co-regulates flowering through its interaction with BBX4/COL3, which may be an additional regulatory mechanism affecting reproductive development in Arabidopsis [15]. In rice, OsCOL3 negatively modulates the expression of Hd3a and TFL under short-day conditions to regulate the flowering time [16]. OsCOL9 can delay the flowering time by inhibiting the Ehd1 pathway, which is consistent with the results of the research on Arabidopsis [17]. Recent research has demonstrated that the chrysanthemum gene CmBBX8 controls the flowering time by activating the CmFTL promoter [18].

According to the research, rapeseed and Arabidopsis are both members of the Brassicaceae family and may share a common ancestor [19,20]. Arabidopsis and rapeseed homologous genes share a high sequence similarity and may also have similar functions [21]. Until recently, there have been few reports on the flowering time of rapeseed. Previous research on the rapeseed flowering time has mostly focused on QTL linkage mapping and several QTL links to the flowering time have been discovered, primarily on chromosomes A02, A09, A10, C02, and C03 [22,23,24]. Genome-wide association studies (GWASs) have recently been used for the genetic dissection of complex features in plants such as rice [25], maize [26], and rapeseed [27,28,29]. A substantial number of single-nucleotide polymorphism (SNP) sites linked to the flowering time in rapeseed have been identified using GWAS analyses. A GWAS analysis of 523 natural populations under 8 distinct environmental conditions revealed that 41 SNPs were strongly associated with the flowering time [28]. Wu et al. found SNPs in the promoter regions of the FT and FLC (FLOWERING LOCUS C) orthologs correlating to distinct rapeseed ecotypes by resequencing 991 rapeseed accessions globally and conducting a GWAS analysis [29]. These studies may provide important molecular markers for accelerating Brassica napus breeding.

This study is novel because it used the CRISPR/Cas9 system to edit four BnaCOL9 homologs in rapeseed and generated a quadruple mutant of BnaCOL9, which has not been previously reported on to the best of our knowledge. Our results indicated that the application of CRISPR/Cas9 technology to create early-flowering germplasm resources provided useful materials to improve the early-maturing breeding of rapeseed.

2. Results

2.1. Isolation of BnaCOL9 from Brassica napus

Using the protein sequence of Arabidopsis AtCOL9, four Brassica napus orthologs of COL9—BnaA0529980D, BnaC0544310D, BnaA0330130D, and BnaC0335440D—were identified and named BnaA05COL9, BnaC05COL9, BnaA03COL9, and BnaC03COL9, respectively. The amino acid sequence identities with Arabidopsis AtCOL9 were 79.63%, 79.85%, 72.95%, and 72.24%, respectively. These results suggested that the BnaCOL9 homologs could have functions similar to those of AtCOL9. A phylogenetic analysis showed that the BnaA05COL9 and BnaA03COL9 proteins were most closely related to the homologs of B. rapa. The BnaC05COL9 and BnaC03COL9 proteins were most closely related to their homologs in B. oleracea (Figure 1A). A MEME analysis showed that BnaA05COL9, BnaA03COL9, BnaC05COL9, and BnaC03COL9 contained ten conserved motifs (Figure 1B). In addition, the amino acid sequences of the four BnaCOL9 homologs shared an 83.29% identity. The B-BOX domain was located at the N-terminal and the CCT domain was located at the C-terminal (Figure 1C). Therefore, these four BnaCOL9 proteins could have redundant functions.

2.2. Knockout of BnaCOL9 Shows an Early-Flowering Phenotype

Brassica napus is an allotetraploid species, which suggests that it has at least two copies of each gene. Brassica napus has four copies of BnaCOL9 and the biological functions may be redundant. To analyze the biological function of BnaCOL9, we designed two sgRNAs to knock out BnaA05COL9, BnaC05COL9, BnaA03COL9, and BnaC03COL9 in rapeseed. sgRNA1 (sg1) was designed on the second exon of BnaA05COL9, BnaC05COL9, and BnaC03COL9; it was on the third exon of BnaA03COL9. sgRNA2 (sg2) was used to target the fourth exon of BnaA05COL9, BnaC05COL9, and BnaC03COL9 as well as the fifth exon of BnaA03COL9 (Figure S1). In total, we obtained 40 T0 regenerated transgenic plants and 38 of them were positive. The primers (Cas9-F and Cas9-R) used to detect the positive plants are listed in Table S1. The percentage of positive plants was 95% (Figure S2). Among the T1 regeneration lines, three independent mutant lines (T1-cr6, T1-cr7, and T1-cr9) in which four copies of BnaCOL9 were simultaneously edited were selected for further research (Figure S3). In the T2 generation, we used Hi-TOM sequencing to confirm the presence of targeted mutations on the target site (Table S2); most of the edit types were single-base insertions. The types of the base deletion mutations, including single-base, four-base, and five-base deletions, were also detected (Figure 2A). The flowering time was assessed and compared with that of the wild-type. Notably, the mutant lines showed earlier flowering times than the wild-type plants (Figure 2B). The average flowering time of the mutant lines, including T2-cr6, T2-cr7, and T2-cr9, was 34.67 ± 1.97 d, 35.16 ± 1.94 d, and 35.33 ± 1.94 d, respectively. In comparison, the average flowering time of the wild-type was 43 ± 1.26 d (Figure 2C). This result indicated that the knockout of BnaCOL9 resulted in an early flowering.

2.3. Overexpression of BnaCOL9 Delays the Flowering Time

To confirm that BnaCOL9 was involved in regulating the flowering time in rapeseed, we generated BnaCOL9-mCherry OE lines in a Westar genetic background and assessed their flowering time. The BnaCOL9 overexpression lines showed a late flowering time compared with the wild-type plants (Figure 3A). The average flowering time of the overexpression lines OE-1, OE-3, and OE-4 was 52 ± 2.28 d, 51.5 ± 1.60 d, and 51.8 ± 1.95 d, respectively, which was significantly later than that of the wild-type (Figure 3B). The qRT-PCR results showed that the expression of BnaA05COL9 was significantly increased in the overexpression lines (Figure 3C). Collectively, these results suggested that BnaCOL9 regulates flowering in rapeseed.

2.4. BnaCOL9 Expression Pattern Analysis and Subcellular Localization

To study the expression pattern of BnaCOL9 in rapeseed, a qRT-PCR analysis was performed using RNA extracted from the roots, leaves, stems, flowers, flower buds, and siliques. The results showed that BnaA05COL9 and BnaC05COL9 had similar expression patterns, which were highest in the leaves, followed by the roots. The transcript of BnaA05COL9 was not detected in the siliques. The main expression tissues differed between BnaA03COL9 and BnaC03COL9. BnaA03COL9 was mainly expressed in the flowers. BnaC03COL9 was mainly expressed in the leaves, followed by the flowers (Figure 4A). To further examine the expression pattern of BnaA05COL9, a construct harboring the GUS gene driven by the promoter of BnaA05COL9 derived from Westar was transformed into rapeseed. Strong GUS signals were detected in the leaves and only weak GUS signals were detected in the roots, flowers, and buds (Figure 4B–E). However, we observed no GUS signals in the siliques or stems. The GUS expression pattern was consistent with the results of the qRT-PCR analysis.

The BnaCOL9 protein was transiently transformed into tobacco plants through an Agrobacterium-mediated technique to determine its subcellular localization. The four BnaCOL9-GFP proteins were all localized in the nucleus (Figure 5), indicating that BnaCOL9 is a nuclear protein. This was consistent with previous reports of it acting as a transcription factor to regulate various biological processes.

2.5. Transcription of Flowering-Related Genes Is Regulated by BnaCOL9

COL9 has been found to directly regulate the flowering-related genes in Arabidopsis plants. The current study used qRT-PCR to detect the floral-related gene expression, including that of BnaFT, BnaCO, and BnaSOC1. The expression levels of these genes were measured in the leaves of the 28-day-old seedlings of the wild-type, BnaCOL9 mutant, and BnaCOL9 overexpression lines (Figure 6). In the quadruple mutant, the expression levels of BnaFT and BnaCO were significantly higher than those in the wild-type. However, the expression of BnaSOC1 did not change. In contrast, the expression levels of BnaFT and BnaCO were significantly reduced in the overexpression lines. The expression of BnaSOC1 remained unchanged. This was consistent with previous reports that COL9 controls the flowering time by inhibiting the FT and CO expression in Arabidopsis. In conclusion, BnaCOL9 might regulate the flowering time by inhibiting the expression of the BnaFT and BnaCO genes.

3. Discussion

The CRISPR/Cas9 gene editing system plays a vital role in studying gene functions and improving crop agronomic traits. Since the first successful application of this technology in plants [30], it has been applied to rice [31], wheat [32], soybeans [33], maize [34], and Brassica napus [35]. For example, a homozygous Brassica napus Bnasvp mutant was obtained using the CRISPR/Cas9 system and resulted in the stable transformation of Brassica napus. It was found that the Bnasvp mutant showed an earlier flowering time under two environmental conditions (summer and winter growing seasons) [36]. Brassica napus with multilocular siliques was obtained by editing the BnaCLV gene using the CRISPR/Cas9 system [35]. In the present study, two synthetic single-guide RNAs (sg1 and sg2) were created with the intention of editing four conserved areas in the BnaCOL9 genes. Through gene editing assays, we found that most of the mutations caused by sg1 and sg2 to the four copies of BnaCOL9 were single-base insertions (Figure 2A). This could lead to a frameshift mutation in the BnaCOL9 protein.

The flowering time is a complex biological process for plants, and research has revealed that approximately 300 genes in Arabidopsis are involved in the flowering regulation [37]. Although an overexpression of COL9 in Arabidopsis has been shown to delay flowering [10], to the best of our knowledge, no reports have been published on whether COL9 can affect the flowering of Brassica napus. We obtained mutants of BnaCOL9 through gene editing and demonstrated that BnaCOL9 could advance the flowering time (Figure 2), but an overexpression of BnaCOL9 could delay the flowering time (Figure 3). By detecting the expression levels of the flowering-related genes in the overexpression and knockout mutants, the expression levels of CO and FT in the knockout mutants were found to be significantly increased whereas those in the overexpression lines were significantly decreased (Figure 6). This indicated that BnaCOL9 might influence the flowering time through the regulation of CO and FT directly or indirectly in rapeseed. At present, however, we have no direct evidence to explain how BnaCOL9 regulates flowering in Brassica napus; this will be the subject of our future work.

An important objective in breeding is to reduce the length of the vegetative stage of crops without significantly reducing the crop yields. Our study showed that certain CRISPR/Cas9 mutations may significantly enhance the agronomic features of Brassica napus. Mutants with earlier flowering times could be created by altering multiple copies of BnaCOL9. By modifying the genes associated with the flowering time, we could produce mutants with various flowering times, giving us more possibilities for early-maturing breeding in Brassica napus.

4. Materials and Methods

4.1. Plant Materials

For the transformation experiments, a spring-type rapeseed cultivar (Westar) was used as the receptor material. We obtained seeds from the National Rapeseed Engineering and Technology Research Center in Wuhan, China. The transgenic and wild-type plants were planted in a greenhouse (16 h light/8 h dark at 22 °C).

4.2. Construction of Phylogenetic Tree and Sequence Alignment

The Arabidopsis AtCOL9 amino acid sequence was submitted to the Darmor-bzh database (https://www.genoscope.cns.fr/brassicanapus/, last access date: 24 November 2022) to search for BnaCOL9 homologs in the rapeseed genome. The following species had full-length protein sequences of BnaCOL9 and its homologs: B. rapa, Brassica oleracea, Ziziphus jujuba, Rosa chinensis, Corchorus olitorius, Carica papaya, Theobroma cacao, Glycine max, Rhamnella rubrinervis, and Arabidopsis. We retrieved these sequences from the NCBI database (https://www.ncbi.nlm.nih.gov/, last access date: 24 November 2022) and constructed a phylogenetic tree to study their evolutionary relationships. MEGA Ⅹ was used to build adjacency unrooted trees [38]. Protein-conserved motifs from COL9 homologs were identified using MEME [39] and TBtools software (version 1.0987650) [40]. An alignment of multiple sequences was performed using DNAMAN software (version 6.0 Lynnon Biosoft, San Ramon, CA, USA).

4.3. Construction of a CRISPR/Cas9 Vector

We designed two sgRNAs using the CRISPR-P2.0 tool (http://crispr.hzau.edu.cn/CRISPR2/, last access date: 24 November 2022) to generate the BnaCOL9 quadruple mutants. A CRISPR/Cas9 binary vector set was provided by Prof. Chen Qijun (China Agricultural University) that included PCBC-D1T2 (with a chloramphenicol resistance) and PKSE401 (with a kanamycin resistance) vectors. The Zea mays Cas9 protein was driven by the double 35S promoter and the two sgRNAs were driven by Arabidopsis promoters U6-26 and U6-29. They were fused to the T-DNA region of the PKSE401 vector. The vector was constructed as previously described [41]. We performed the experiments following the improved Agrobacterium-mediated genetic transformation method of Dai et al. [42]. The transformation steps were as follows: 1. First, we sowed seeds of sterilized rapeseed on an M0 medium and cultured them in a dark room for 7 days. The hypocotyl was picked out (the cotyledons and roots were removed). 2. We prepared the infested Agrobacterium bacterium solution. 3. We prepared the co-culture. 4. We performed the callus induction. The explants were cultured on the medium of M2 for 21 days. 5. We performed the shoot differentiation. The explants on M2 were transferred to the medium of M3 for 14 days; after that, the explants were transferred to the new medium of M3 every 14 days until the shoots came out. 6. We checked the root system for the initiation and transplantation of the seedlings. The shoots were cut from the junctions of the explants and transplanted to a clear square box of M4 for the culture. Usually, it took 2–3 weeks for the roots to grow.

4.4. Transgenic Plant Mutation Site Identification

We used a high-throughput sequencing platform to detect the mutant sites in the transgenic plants and submitted the sequencing results to the Hi-TOM decoding website to visualize the mutant sites [43]. The specific steps were as follows: 1. Target amplification using specific primers (first round of PCR, Table S1). 2. Differential labeling of each plant with universal primers (second round of PCR). 3. Aliquots of the second round PCR products were mixed and subjected to next-generation sequencing (Annoroad Gene Technology, Beijing, China). 4. The sequencing results were decoded using an online decoding tool (http://www.hi-tom.net/hi-tom/, last access date: 24 November 2022) to show the mutation status of the target site.

4.5. RNA Extraction and qRT-PCR

RNA was extracted from the roots, stems, leaves, flowers, buds, and siliques of Westar to measure the expression of BnaCOL9. The total RNA from the transgene plants was extracted to detect the transcript levels of BnaCOL9, BnaFT, BnaCO, and BnaSOC1. Total RNA was extracted from the tissue samples using an RNA prep plant kit (Promega, Shanghai, China) containing a DNase I treatment reagent after harvesting and freezing in liquid nitrogen. First-strand cDNA was reverse-transcribed using a Transcript RT Kit (R223, Vazyme, Nanjing, China). For the qRT-PCR analysis, three biological replicates were used. According to the manufacturer’s protocol, the relative gene expression was determined using SYBR qPCR Master Mix (Q311, Vazyme) with a CFX384 touch real-time PCR detection system (Bio-Rad, Hercules, CA, USA). The 2^−ΔΔCt method was used to calculate the relative expression levels [44]. The mean values for each measured parameter were compared using two-tailed, two-sample Student’s t-tests in GraphPad Prism (version 8.0, GraphPad Software, San Diego, CA, USA) as appropriate.

4.6. Subcellular Localization

The coding sequences of BnaAO5COL9, BnaCO5COL9, BnaAO3COL9, and BnaC03COL9 were amplified without the termination codon using specific primers (Table S1). To generate the C-terminal GFP fusion products, the sequences were inserted into the transient expression vector PH7Lic6.0-35S-GFP. The constructed fusion plasmid of green fluorescent protein was transferred into Agrobacterium GV3101. The transformed cells were harvested and resuspended in 10 mM MES-KOH (pH 5.6) containing 10 mM MgCl₂ and 150 mM acetosyringone and then mixed. The final optical density (OD600) at 600 nm was 0.8. After incubation for 2 h at 22 °C, the Agrobacterium suspension was injected into the expanded leaves of 4-week-old tobacco plants (Nicotiana benthamiana). The fluorescence signals in the tobacco leaves were imaged using a confocal laser scanning system (Leica SP8 DLS, Germany) [45].

4.7. GUS Assay

A 1.6 kb upstream sequence from the translation start site of BnaAO5COL9 was cloned from Damor- bzh. This sequence was cloned into the pCAMBIA2300 binary vector. As previously described, the pBnaA05COL9-GUS fusion construct was introduced into Westar via an Agrobacterium-mediated transformation [42]. Positive transgenic plants were identified using the primers M13-47 and GUS-R (Table S1). The GUS activity was determined by staining different tissues from positive transgenic plants in an X-GLU solution overnight at 37 °C. The tissues were then cleared with 75% (v/v) ethanol until the pigments were completely removed.