CRISPR/Cas9-Mediated Mutagenesis of BrLEAFY Delays the Bolting Time in Chinese Cabbage (Brassica rapa L. ssp. pekinensis)

Chinese cabbage has unintended bolting in early spring due to sudden climate change. In this study, late-bolting Chinese cabbage lines were developed via mutagenesis of the BrLEAFY (BrLFY) gene, a transcription factor that determines floral identity, using the clustered regularly interspaced short palindromic repeat (CRISPR)/CRISPR-associated protein 9 (CRISPR/Cas9) system. Double-strand break of the target region via gene editing based on nonhomologous end joining (NHEJ) was applied to acquire useful traits in plants. Based on the ‘CT001’ pseudomolecule, a single guide RNA (sgRNA) was designed and the gene-editing vector was constructed. Agrobacterium-mediated transformation was used to generate a Chinese cabbage line in which the sequence of the BrLFY paralogs was edited. In particular, single base inserted mutations occurred in the BrLFY paralogs of the LFY-7 and LFY-13 lines, and one copy of T-DNA was inserted into the intergenic region. The selected LFY-edited lines displayed continuous vegetative growth and late bolting compared to the control inbred line, ‘CT001’. Further, some LFY-edited lines showing late bolting were advanced to the next generation. The T-DNA-free E1 LFY-edited lines bolted later than the inbred line, ‘CT001’. Overall, CRISPR/Cas9-mediated mutagenesis of the BrLFY gene was found to delay the bolting time. Accordingly, CRISPR/Cas9 is considered an available method for the molecular breeding of crops.


Introduction
In Chinese cabbage, bolting, which appears as flower buds from leaves, is controlled by internal signals and environmental factors. Internal signals have been investigated in various pathways, including autonomous circadian rhythm and floral meristem identity [1][2][3]. In particular, the transcription factors (TFs), LEAFY (LFY), APETALA1 (AP1), and CAULIFLOWER (CAL), and the floral meristem identity pathway control flower development and flowering time [4,5]. Functional analysis of these genes was mainly performed to determine the morphology of flower organs and assess meristem formation. For example, lfy/ap double mutants showed complete transformation of all whorled flowers into inflorescence [6]. Floral meristem identity genes interact together to result in morphologic constructions, including short internodes and suppression of axillary buds.
Temperature is a typical environmental factor that induces bolting. Chinese cabbage is a seed vernalization-type plant. After a certain period of response to low temperatures, and the external temperature suitable for growth increases, Chinese cabbage displays bolting and flowering [7]. Vernalization is usually necessary to promote flowering; however, recent rapid changes in temperatures due to climate change have sometimes led to undesirable premature bolting, which affects sufficient vegetative growth for leaf formation and causes commercial quality degradation in vegetable crops [8,9]. Premature bolting and flowering of leafy plants like lettuce, radish, and Chinese cabbage are unfavorable agricultural traits, resulting in significant economic loss. Chinese cabbage is a leafy vegetable that is widely

Target Gene Structure Analysis and LFY-Editing Vector Construction
To construct an LFY-editing vector, orthologs of the BrLFY gene were analyzed based on the AtLFY (AT5G61850) coding DNA sequence using the B. rapa database (BRAD, http://brassicadb.org, accessed on 8 April 2019). Based on the similar sequences obtained, two paralogs (CT001_A02071870 and CT001_A06214910) were identified in the Chinese cabbage genome. The amino acid sequence similarity between the BrLFY paralogs was 90%. CT001_A02071870 was identified on the A02 chromosome while CT001_A06214910 was identified on the A06 chromosome. Both genes are composed of three exons spanning 2.5-2.8 kb. The single guide RNA (sgRNA) was designed simultaneously with target paralogs and included a partial sequence of the exonic region containing the NGG sequence (PAM sequence). In the third exon, the DNA-binding domain of the BrLFY gene was selected as the target region ( Figure 1). The off-targeting effect was analyzed, and an LFY-editing vector (pLFY) carrying sgRNA, sgRNA scaffold, and Cas9 components was constructed. pLFY was modified based on the pHAtC gene editing vector, which contained hygromycin resistance as a selection marker and transformed into Agrobacterium tumefaciens LBA4404.

Target Gene Structure Analysis and LFY-Editing Vector Construction
To construct an LFY-editing vector, orthologs of the BrLFY gene were analyzed based on the AtLFY (AT5G61850) coding DNA sequence using the B. rapa database (BRAD, http://brassicadb.org, accessed on 8 April 2019). Based on the similar sequences obtained, two paralogs (CT001_A02071870 and CT001_A06214910) were identified in the Chinese cabbage genome. The amino acid sequence similarity between the BrLFY paralogs was 90%. CT001_A02071870 was identified on the A02 chromosome while CT001_A06214910 was identified on the A06 chromosome. Both genes are composed of three exons spanning 2.5−2.8 kb. The single guide RNA (sgRNA) was designed simultaneously with target paralogs and included a partial sequence of the exonic region containing the NGG sequence (PAM sequence). In the third exon, the DNA-binding domain of the BrLFY gene was selected as the target region ( Figure 1). The off-targeting effect was analyzed, and an LFY-editing vector (pLFY) carrying sgRNA, sgRNA scaffold, and Cas9 components was constructed. pLFY was modified based on the pHAtC gene editing vector, which contained hygromycin resistance as a selection marker and transformed into Agrobacterium tumefaciens LBA4404.

Selection of Tentative LFY-Edited Chinese Cabbage Lines Using PCR Analysis
To develop the LFY-edited lines, seeds of the inbred line, 'CT001' were germinated in vitro and re-differentiation of shoots derived from hypocotyls was induced by Agrobacterium transformation. Fourteen independent tentative E 0 LFY-edited lines were obtained. To analyze the gene mutations induced by CRISPR/Cas9, DNA was extracted from leaves to select the T-DNA-inserted lines. Polymerase chain reaction (PCR) analysis was performed using two primer sets that amplify a part of Hyg R and Cas9hc:NLS:HA (Cas9hc) present in the T-DNA (Table S1 in Supplementary Materials). Nine of the 14 E 0 lines appeared as target amplicons ( Figure 2).

Selection of Tentative LFY-Edited Chinese Cabbage Lines using PCR Analysis
To develop the LFY-edited lines, seeds of the inbred line, 'CT001' were germina in vitro and re-differentiation of shoots derived from hypocotyls was induced Agrobacterium transformation. Fourteen independent tentative E0 LFY-edited lines w obtained. To analyze the gene mutations induced by CRISPR/Cas9, DNA was extrac from leaves to select the T-DNA-inserted lines. Polymerase chain reaction (PCR) analy was performed using two primer sets that amplify a part of Hyg R and Cas9hc:NLS:H (Cas9hc) present in the T-DNA (Table S1 in Supplementary Materials). Nine of the 14 lines appeared as target amplicons ( Figure 2).

Identification of CRISPR/Cas9-Mediated Mutagenesis of the BrLFY Paralogs
Nine independent E0 LFY-edited lines with target amplicons were analyzed for tar sequence mutations. As a result, four E0 LFY-edited lines showed a single base insert in CT001_A02071870 and three E0 LFY-edited lines showed a single base insertion CT001_A06214910 ( Figure S1 in Supplementary Materials). Among the selected LF edited lines with a single base insertion, the E0 LFY-edited lines (LFY-7 and LFY-13) w identified again via PCR and reverse-transcript PCR (RT-PCR) analysis using target ge specific amplification primer sets (Table S1 in Supplementary Materials; Figure 3 Sequence confirmation of the target region in the two paralogs was performed, a insertion mutation of a single base was induced by CRISPR/Cas9 ( Figure 3B). An aden base of the target regions near the PAM sequence of sgRNA in BrLFY paralogs w identified in the E0 LFY-7 line. The E0 LFY-13 line was also found to have an insert mutation in the target regions of the BrLFY paralogs. In the CT001_A02071870 gene of E0 LFY-13 line, a single base (adenine) was inserted 4 bp upstream of the PAM sequen In addition, the CT001_A06214910 gene was targeted and a cytosine base was inserte bp upstream of the PAM sequence ( Figure 3C). CRISPR/Cas9-mediated mutagenesis the BrLFY paralogs resulted in a frameshift and premature termination codons.
Then, the expression levels of BrLFY paralogs in the E0 LFY-edited lines w analyzed using quantitative real-time PCR (qPCR) analysis. Overall, the expression le of BrLFY paralogs was lower in the E0 LFY-edited lines than in the inbred line 'CT00 The occurrence of insertion nucleotide mutation in the coding region affected expression levels of BrLFY paralogs ( Figure S2 in Supplementary Materials).

Identification of CRISPR/Cas9-Mediated Mutagenesis of the BrLFY Paralogs
Nine independent E 0 LFY-edited lines with target amplicons were analyzed for target sequence mutations. As a result, four E 0 LFY-edited lines showed a single base insertion in CT001_A02071870 and three E 0 LFY-edited lines showed a single base insertion in CT001_A06214910 ( Figure S1 in Supplementary Materials). Among the selected LFY-edited lines with a single base insertion, the E 0 LFY-edited lines (LFY-7 and LFY-13) were identified again via PCR and reverse-transcript PCR (RT-PCR) analysis using target gene-specific amplification primer sets (Table S1 in Supplementary Materials; Figure 3A). Sequence confirmation of the target region in the two paralogs was performed, and insertion mutation of a single base was induced by CRISPR/Cas9 ( Figure 3B). An adenine base of the target regions near the PAM sequence of sgRNA in BrLFY paralogs was identified in the E 0 LFY-7 line. The E 0 LFY-13 line was also found to have an insertion mutation in the target regions of the BrLFY paralogs. In the CT001_A02071870 gene of the E 0 LFY-13 line, a single base (adenine) was inserted 4 bp upstream of the PAM sequence. In addition, the CT001_A06214910 gene was targeted and a cytosine base was inserted 4 bp upstream of the PAM sequence ( Figure 3C). CRISPR/Cas9-mediated mutagenesis of the BrLFY paralogs resulted in a frameshift and premature termination codons.
Then, the expression levels of BrLFY paralogs in the E 0 LFY-edited lines were analyzed using quantitative real-time PCR (qPCR) analysis. Overall, the expression level of BrLFY paralogs was lower in the E 0 LFY-edited lines than in the inbred line 'CT001'. The occurrence of insertion nucleotide mutation in the coding region affected the expression levels of BrLFY paralogs ( Figure S2 in Supplementary Materials).

Observation of the Bolting Time of the Inbred Line, 'CT001', and E 0 LFY-Edited Lines
To determine whether the BrLFY gene was related to bolting, inbred line, 'CT001', and E 0 LFY-edited lines were planted in a pot and placed in a cold room at KyungHee Univ. After vernalization, the inbred line, 'CT001' and LFY-7, and LFY-13 edited lines were transferred to a greenhouse. Normal growth and development were observed for all three lines. A difference in bolting and flowering time was found between the inbred line, 'CT001', and LFY-edited lines. Further, in the initial stage, the LFY-edited lines had lower heights and more leaves. The first flower bud appeared in 'CT001'. However, bolting was delayed in LFY-7 and LFY-13 lines by approximately 7 days compared to that in 'CT001' (Figure 4). There was no difference in the flower bud shape between 'CT001' and the E 0 LFY-edited lines. These results indicate that the mutation of the BrLFY gene in the LFY-edited lines resulted in delayed bolting in Chinese cabbage.

Observation of the Bolting Time of the Inbred Line, 'CT001', and E0 LFY-Edited Lines
To determine whether the BrLFY gene was related to bolting, inbred line, 'CT001', and E0 LFY-edited lines were planted in a pot and placed in a cold room at KyungHee Univ. After vernalization, the inbred line, 'CT001' and LFY-7, and LFY-13 edited lines were transferred to a greenhouse. Normal growth and development were observed for all three lines. A difference in bolting and flowering time was found between the inbred line, 'CT001', and LFY-edited lines. Further, in the initial stage, the LFY-edited lines had lower heights and more leaves. The first flower bud appeared in 'CT001'. However, bolting was delayed in LFY-7 and LFY-13 lines by approximately 7 days compared to that in 'CT001' (Figure 4). There was no difference in the flower bud shape between 'CT001' and the E0 LFY-edited lines. These results indicate that the mutation of the BrLFY gene in the LFYedited lines resulted in delayed bolting in Chinese cabbage.

T-DNA Copy Number and Site Analysis of the E0 LFY-Edited Lines
The T-DNA copy number was confirmed using Southern analysis, probing with the hpt gene. A 709 bp probe was designed and radiolabeled with 32 P. The site was determined using the λ HindIII molecular ladder markers on the blot. Based on the results, LFY-7 and LFY-13 contained a single copy insertion ( Figure 5A). No signal was observed in the negative control line, 'CT001'. To determine whether T-DNA insertion affected other gene functions, the junction regions between the Chinese cabbage genome and T-DNA border were amplified by variable argument thermal asymmetric interlaced PCR (VA-TAIL PCR) analysis, as previously described. Using six arbitrary degenerate (AD) primers, at least one fragment was amplified in the junction region. Among the examined amplicons, two contained specific junction sequences.
T-DNA was stably inserted into the intergenic region of the LFY-7 and LFY-13 edited lines, including the insertion of some nucleotide sequences, without affecting the other gene functions ( Figure 5B). In the LFY-7 edited line, T-DNA was inserted in the intergenic region of the CT001_A09354000 and CT001_A09354010. T-DNA of the LFY-13 edited line was located on the intergenic region of the CT001_A04131960 and CT001_A04131970.

T-DNA Copy Number and Site Analysis of the E 0 LFY-Edited Lines
The T-DNA copy number was confirmed using Southern analysis, probing with the hpt gene. A 709 bp probe was designed and radiolabeled with 32 P. The site was determined using the λ HindIII molecular ladder markers on the blot. Based on the results, LFY-7 and LFY-13 contained a single copy insertion ( Figure 5A). No signal was observed in the negative control line, 'CT001'. To determine whether T-DNA insertion affected other gene functions, the junction regions between the Chinese cabbage genome and T-DNA border were amplified by variable argument thermal asymmetric interlaced PCR (VA-TAIL PCR) analysis, as previously described. Using six arbitrary degenerate (AD) primers, at least one fragment was amplified in the junction region. Among the examined amplicons, two contained specific junction sequences.

T-DNA Copy Number and Site Analysis of the E0 LFY-Edited Lines
The T-DNA copy number was confirmed using Southern analysis, probing with the hpt gene. A 709 bp probe was designed and radiolabeled with 32 P. The site was determined using the λ HindIII molecular ladder markers on the blot. Based on the results, LFY-7 and LFY-13 contained a single copy insertion ( Figure 5A). No signal was observed in the negative control line, 'CT001'. To determine whether T-DNA insertion affected other gene functions, the junction regions between the Chinese cabbage genome and T-DNA border were amplified by variable argument thermal asymmetric interlaced PCR (VA-TAIL PCR) analysis, as previously described. Using six arbitrary degenerate (AD) primers, at least one fragment was amplified in the junction region. Among the examined amplicons, two contained specific junction sequences.
T-DNA was stably inserted into the intergenic region of the LFY-7 and LFY-13 edited lines, including the insertion of some nucleotide sequences, without affecting the other gene functions ( Figure 5B). In the LFY-7 edited line, T-DNA was inserted in the intergenic region of the CT001_A09354000 and CT001_A09354010. T-DNA of the LFY-13 edited line was located on the intergenic region of the CT001_A04131960 and CT001_A04131970.  T-DNA was stably inserted into the intergenic region of the LFY-7 and LFY-13 edited lines, including the insertion of some nucleotide sequences, without affecting the other gene functions ( Figure 5B). In the LFY-7 edited line, T-DNA was inserted in the intergenic region of the CT001_A09354000 and CT001_A09354010. T-DNA of the LFY-13 edited line was located on the intergenic region of the CT001_A04131960 and CT001_A04131970.

Inheritance of Base Mutation in the E 1 T-DNA-Free LFY-Edited Lines
The E 1 LFY-edited lines were analyzed to identify the target sequence mutations.
The T-DNA-free E 1 LFY-edited lines were confirmed by PCR analysis and partial lines were selected using both primer sets ( Figure S3 in Supplementary Materials). Three of the 12 LFY-7 and one of the five LFY-13 edited lines were selected. Target sequence mutation analysis was conducted using PCR analysis with the target gene-specific amplifying primer sets (Table S1 in Supplementary Materials). Base mutations induced by CRISPR/Cas9 in the two paralogs were stably inherited by the next generation. Genetic mutations were compared between E 0 and E 1 LFY-edited lines. Some E 1 LFY-edited lines with T-DNA insertions retained the same mutated sequence as the E 0 LFY-edited lines. However, some did not inherit mutated sequences and showed other sequence mutations. The E 1 #7-1 line with T-DNA insertion inherited the mutation pattern of CT001_A02071870 and CT001_A06214910 genes, but the E 1 #7-4 line with T-DNA insertion showed an altered mutation pattern with cytosine base inserted in the CT001_A06214910 gene. Similarly, the E 1 #13-3 line with T-DNA insertion showed a mutation pattern in which one base was inserted in the CT001_A02071870 and CT001_A06214910 genes.
Among the progeny lines of LFY-7, the same base mutation pattern was found in the T-DNA-free E 1 #7-2 and #7-9 lines ( Figure 6). An adenine base was inserted in both paralogs, and owing to this mutation, a premature termination was predicted to appear. Likewise, the E 1 #13-1 progeny line of the T-DNA-free LFY-13 line, CT001_A02071870, showed insertion of an adenine base, while CT001_A06214910 showed insertion of a cytosine base.

Inheritance of Base Mutation in the E1 T-DNA-Free LFY-Edited Lines
The E1 LFY-edited lines were analyzed to identify the target sequence mutations. T T-DNA-free E1 LFY-edited lines were confirmed by PCR analysis and partial lines we selected using both primer sets ( Figure S3 in Supplementary Materials). Three of the LFY-7 and one of the five LFY-13 edited lines were selected. Target sequence mutati analysis was conducted using PCR analysis with the target gene-specific amplifyi primer sets (Table S1 in Supplementary Materials). Base mutations induced CRISPR/Cas9 in the two paralogs were stably inherited by the next generation. Gene mutations were compared between E0 and E1 LFY-edited lines. Some E1 LFY-edited lin with T-DNA insertions retained the same mutated sequence as the E0 LFY-edited lin However, some did not inherit mutated sequences and showed other sequence mutation The E1 #7-1 line with T-DNA insertion inherited the mutation pattern of CT001_A020718 and CT001_A06214910 genes, but the E1 #7-4 line with T-DNA insertion showed an alter mutation pattern with cytosine base inserted in the CT001_A06214910 gene. Similarly, t E1 #13-3 line with T-DNA insertion showed a mutation pattern in which one base w inserted in the CT001_A02071870 and CT001_A06214910 genes.
Among the progeny lines of LFY-7, the same base mutation pattern was found in t T-DNA-free E1 #7-2 and #7-9 lines ( Figure 6). An adenine base was inserted in bo paralogs, and owing to this mutation, a premature termination was predicted to appe Likewise, the E1 #13-1 progeny line of the T-DNA-free LFY-13 line, CT001_A0207187 showed insertion of an adenine base, while CT001_A06214910 showed insertion of cytosine base.

Delayed Bolting in the E1 T-DNA-Free LFY-Edited Lines
To confirm the bolting time and observe the developmental processes, the inbred li 'CT001' and E1 T-DNA-free LFY-edited lines were cold-treated in a cold room KyungHee Univ.
The flowering time of the E1 T-DNA-free LFY-edited lines was later than that of t control line, 'CT001' (Figure 7A). When bolting was observed in the inbred line, 'CT00 the E1 LFY-edited lines were still in the vegetative growth stage. Thereafter, when t delay was approximately 10 days on average, the E1 #7-2 and #13-1 LFY-edited lines we converted to the reproductive growth stage, and bolting was observed. The E1 lin displayed a phenotype similar to that of the E0 LFY-edited lines. Except for the delay bolting phenomenon, no differences were found in the developmental processes in

Delayed Bolting in the E 1 T-DNA-Free LFY-Edited Lines
To confirm the bolting time and observe the developmental processes, the inbred line 'CT001' and E 1 T-DNA-free LFY-edited lines were cold-treated in a cold room at KyungHee Univ.
The flowering time of the E 1 T-DNA-free LFY-edited lines was later than that of the control line, 'CT001' (Figure 7A). When bolting was observed in the inbred line, 'CT001', the E 1 LFY-edited lines were still in the vegetative growth stage. Thereafter, when the delay was approximately 10 days on average, the E 1 #7-2 and #13-1 LFY-edited lines were converted to the reproductive growth stage, and bolting was observed. The E 1 lines displayed a phenotype similar to that of the E 0 LFY-edited lines. Except for the delayed bolting phenomenon, no differences were found in the developmental processes in all lines. The delay in bolting time was confirmed to be caused by CRISPR/Cas9-mediated mutagenesis of the BrLFY gene.

Discussion
TFs possess reprogramming activity that enables changes from one cell type another. The BrLEAFY (BrLFY) gene is a TF that determines flower identity and has t domains. A sterile alpha motif (SAM) is located at the N-terminal amino acid sequence the BrLFY gene, while a DNA-binding domain (DBD) is located at the C-terminal am acid sequence [23]. The DNA-binding domain is included in a protein involved in regulation of gene expression and can recognize a specific DNA sequence or hav general affinity for DNA. The BrLFY gene has a unique helix-turn-helix struct observed in proteins that regulate developmental processes [24]. During developmental processes of floral plants, the transition from vegetative to reproduct growth occurs.
The LFY gene directly or indirectly interacts with another floral identity ge APETALA1 (AP1) [22]. The downstream network of LFY controls a feed-forward react that regulates the upregulation of AP1. LFY is also associated with gibberellin, a hormo involved in flowering [25]. Increasing the LFY activity decreases gibberellin levels, wh increases the levels of GA-GID1-DELLA proteins (DELLAs). Flowering is promoted wh DELLAs are decomposed; however, owing to the decrease in gibberellin levels, DELL cannot be degraded, which affects the delay in flowering [26].
Previous studies revealed that the LFY gene acts as a switch that induces floweri Overexpression of LFY induces early flowering in Arabidopsis transgenic plants [27] transgenic Chinese cabbage plant developed by RNA interference-media transformation displayed late bolting and flowering due to the downregulation of L gene expression [28]. Transgenic plants were found to continue to grow in the vegetat stage. Further, the number of leaves was greater than that of the control, and the pl stem length was shorter than that of the control owing to the delayed transition reproductive growth. These studies indicate that changes in LFY levels affect flower times.
The BrLFY transcript was found to be present in all Chinese cabbage tissues. It w primarily expressed in young leaves, moderately in older leaves, and slightly in roots a flowers [21]. Similarly, in this study, the expression of the BrLFY paralogs was confirm in tissues such as leaves, flowers, stems, and flower buds of the inbred line 'CT001' (Fig  S4 in Supplementary Materials). In comparison to the shoot apical meristem and ot tissues in Arabidopsis, the LFY gene expression was higher in young and mature lea [29]. In Brassica juncea, mature leaves showed the highest level of BjLFY expressi followed by stems and bracts [30]. In contrast, it was observed that the AFL1 gene of ap and the FaLFY gene of strawberry were not expressed in leaves [31,32].

Discussion
TFs possess reprogramming activity that enables changes from one cell type to another. The BrLEAFY (BrLFY) gene is a TF that determines flower identity and has two domains. A sterile alpha motif (SAM) is located at the N-terminal amino acid sequence of the BrLFY gene, while a DNA-binding domain (DBD) is located at the C-terminal amino acid sequence [23]. The DNA-binding domain is included in a protein involved in the regulation of gene expression and can recognize a specific DNA sequence or have a general affinity for DNA. The BrLFY gene has a unique helix-turn-helix structure observed in proteins that regulate developmental processes [24]. During the developmental processes of floral plants, the transition from vegetative to reproductive growth occurs.
The LFY gene directly or indirectly interacts with another floral identity gene, APETALA1 (AP1) [22]. The downstream network of LFY controls a feed-forward reaction that regulates the upregulation of AP1. LFY is also associated with gibberellin, a hormone involved in flowering [25]. Increasing the LFY activity decreases gibberellin levels, which increases the levels of GA-GID1-DELLA proteins (DELLAs). Flowering is promoted when DELLAs are decomposed; however, owing to the decrease in gibberellin levels, DELLAs cannot be degraded, which affects the delay in flowering [26].
Previous studies revealed that the LFY gene acts as a switch that induces flowering. Overexpression of LFY induces early flowering in Arabidopsis transgenic plants [27]. A transgenic Chinese cabbage plant developed by RNA interference-mediated transformation displayed late bolting and flowering due to the downregulation of LFY gene expression [28]. Transgenic plants were found to continue to grow in the vegetative stage. Further, the number of leaves was greater than that of the control, and the plant stem length was shorter than that of the control owing to the delayed transition to reproductive growth. These studies indicate that changes in LFY levels affect flowering times.
The BrLFY transcript was found to be present in all Chinese cabbage tissues. It was primarily expressed in young leaves, moderately in older leaves, and slightly in roots and flowers [21]. Similarly, in this study, the expression of the BrLFY paralogs was confirmed in tissues such as leaves, flowers, stems, and flower buds of the inbred line 'CT001' (Figure S4 in Supplementary Materials). In comparison to the shoot apical meristem and other tissues in Arabidopsis, the LFY gene expression was higher in young and mature leaves [29]. In Brassica juncea, mature leaves showed the highest level of BjLFY expression, followed by stems and bracts [30]. In contrast, it was observed that the AFL1 gene of apple and the FaLFY gene of strawberry were not expressed in leaves [31,32].
It is known that the LFY gene is involved in flowering by interacting with other genes. FLOWERING LOCUS T (FT) expression directly promotes flowering by regulating and inducing the expression of AP1 and LFY [33]. Overexpression of AGAMOUS-LIKE 17 (AGL17) significantly increased the expression levels of CONSTANS (CO), LFY, and SOC1 and induced early flowering [34]. The B-functional proteins have direct and functional interaction with AP1, LFY, and UNUSUAL FLORAL ORGANS (UFO), as shown by the analysis of the protein binding network between the BrMADS-box proteins [35].
Gene editing studies have been performed using gene scissors that were developed in various forms. Gene editing has been used in research for more than 20 years and has developed rapidly. Research on gene editing began with ZFN, in which Fok1 performs cutting by recognizing a specific DNA sequence using a protein that is structurally stabilized by zinc ions [36]. Since then, Talen, which has improved accuracy owing to the precise recognition of DNA bases, has been developed as a second-generation gene scissor [37]. As a defense mechanism in bacteria, when a virus invades, a DNA fragment is inserted into its genome. A small RNA was prepared and bound to a Cas9 protein. When the virus re-invades, RNA finds and binds to the viral DNA, which is then cut by Cas9 and blocked from replicating [38]. By applying this principle to plants, changes in the nucleotide sequence of a specific gene can be generated. Gene editing studies have been conducted for various animals and plants [39].
Multiple sgRNAs and promoters were configured in one vector to induce the mutagenesis of several homologous genes simultaneously. A single CRISPR/Cas9 editing vector that can edit the ZmTGA family was constructed to generate knockout mutants. To induce the simultaneous mutation of the three homologous genes, two targets were selected based on the conserved sequence [40]. The ZmTGA9 triple mutants failed to produce pollen grains. Homologous genes must be simultaneously targeted to analyze gene function, and an effective sgRNA must be designed.
Using conserved sequences, two to five homologous genes can be targeted using a single sgRNA sequence. The phytoene desaturase (PDS) gene involved in carotenoid biosynthesis was analyzed in Chinese kale (Brassica oleracea var. alboglabra). The CRISPR/Cas9 system was found to generate mutations in two homologous PDS genes using the one sgRNA, which targets the conserved sequence of homologous genes. BaPDS1 and BaPDS2 double mutants at the target site exhibit a clear albino phenotype [41]. Squamosa-promoter binding protein-like 3 (SPL3), a gene predicted to influence the regulation of developmental stage transition, was found to have five homologous genes in the rapeseed (brassica napus) genome. CRISPR/Cas9-mediated mutations were induced by selecting one sgRNA using the consensus sequence of five SPL3 homologous genes. Two individual lines with sequence mutations at all five target sites were identified, and BnSPL3 mutants displayed a developmentally delayed phenotype [42]. These results indicate that one conserved sgRNA was efficient for mutagenesis. In this study, sgRNAs were designed to target BrLFY paralogs using conserved sequences ( Figure 1). Further, sgRNA was designed to target the exonic regions of the BrLFY genes identified to have two copies in the target genome.
CRISPR/Cas9-mediated mutagenesis occurred in the target regions of both BrLFY paralogs (CT001_A02071870 and CT001_A06214910). In this study, mutagenesis of CT001_A02071870 and CT001_A06214910 was performed at the genomic and amino acid levels in the LFY-7 and LFY-13 edited lines. A frameshift was predicted to occur because of the confirmation of one base insertion, which resulted in a change in the amino acid sequence and an early termination codon ( Figure 3). Accordingly, normal LFY protein production was not performed, resulting in delayed bolting relative to that in the inbred line, 'CT001'. Similarly, in plants with one nucleotide inserted into the target sequence of the boltingrelated VERNALIZATION1 (VRN1) gene of Chinese cabbage, bolting and flowering were delayed owing to loss-of-function of the VRN1 gene [12].
The position of the inserted T-DNA in the intergenic region, which does not affect the expression levels of other genes, was confirmed. Subsequently, transgene isolation was carried out using advanced generation progresses. The T-DNA-free E 1 LFY-edited lines bolted later than the inbred line, 'CT001', indicating stable maintenance of CRISPR/Cas9mediated target sequence mutation.
Further, gene editing technology is identified to be distinct from genetically modified organisms (GMOs) as artificial genes are not injected into the genomes of animals and plants, and the existing genes of animals and plants are reconstructed by themselves [43,44]. With the advancement and success of gene editing technology, it is possible to minimize side effects caused by reconstructing the nucleotide sequence via the insertion or deletion of DNA in the base sequence of a specific gene.
Most of the functional analysis of the LFY gene was limited to flower organs, but this research derived novel findings by conducting a study on the bolting phenomenon. Our data also showed that the CRISPR/Cas9 system is effective, and CRISPR/Cas9-mediated mutagenesis of BrLEAFY delays the bolting time in Chinese cabbage (Brassica rapa L. ssp. pekinensis).

Design of Efficient sgRNA and Construction of the Gene Editing Vector
The orthologs of the BrLFY gene were analyzed using the full-length AtLFY (AT5G61850) gene coding DNA sequence from the Arabidopsis Information Resource (TAIR) database (TAIR, https://www.arabidopsis.org/index.jsp, accessed on 8 April 2019). The BrLFY gene sequence was confirmed by matching the obtained sequence to the Brassica rapa genome (BRAD, http://brassicadb.org, accessed on 8 April 2019). Two paralogs of BrLFY (CT001_A02071870 and CT001_A06214910) were present in the genome of the inbred line, 'CT001'. The design of efficient sgRNA sequences and construction of gene editing vector was performed as described previously [45].
sgRNAs with 20 random guide RNA sequences and 3 bp NGG sequences (N can be A, T, G, or C) were designed to target the CT001_A02071870 and CT001_A06214910 genes using CRISPR direct (http://crispr.dbcls.jp; accessed on 10 April 2019). The discovery of off-target cleavage sites was analyzed by applying all possible nucleotide sequence combinations to the NGG sequences of the candidate target sequences. sgRNA was selected as a sequence with as few similar sequences in the Chinese cabbage genome as possible, which showed adequate GC content and targeted the exonic region of a gene. The sgRNA targeting the BrLFY paralogs was named LFY_sgRNA. LFY_sgRNA was ligated to the pHAtC backbone using SacI restriction enzyme, the AtU6 promoter, and scaffold sequence. Clones were confirmed using the BTSeq™ sequencing service (Celemics, Seoul, Republic of Korea). The constructed gene-editing vector was transferred into Agrobacterium tumefaciens LBA4404 and used for Chinese cabbage transformation [46].

Production of the LFY-Edited Chinese Cabbage Lines and Selection Using PCR Analysis
Agrobacterium-mediated transformation of Chinese cabbage was performed [47]. The seeds of the inbred line, 'CT001', were cultured in MS basal medium in vitro and darktreated for hypocotyl elongation. After two weeks of hygromycin selection pressure covering the transient expression period of Agrobacterium, the callus was transferred to a selective medium containing a selection marker. Notably, tentative T-DNA-inserted plants could grow and regenerate. Plants were cultivated in a controlled tissue-culture room at 20 • C under a 16 h light/8 h dark photoperiod.
Stable insertion of the T-DNA was detected via PCR analysis using the genomic DNA of 'CT001' and the tentative LFY-edited lines. Two sets of primers amplified the sections of the Hyg R and Cas9hc regions present in the T-DNA. PCR amplification was performed with 2X H-Star Taq PCR master mix (Biofact, Seoul, Republic of Korea) in a total volume of 30 µL. The following reaction conditions were employed: initial denaturation at 95 • C for 15 min; 35 cycles of 95 • C for 30 s, 60 • C for 20 s, and 72 • C for 45 s; and a final extension step at 72 • C for 5 min. The PCR amplicons were electrophoresed on a 1% agarose.

Analysis of Mutagenesis and Expression Levels of the Target Gene
PCR and RT-PCR analysis was performed to identify the mutation events. The PCR amplicons were eluted using the P&C Multiple Elution Kit (Biosolution, Suwon, Republic of Korea) and sequenced using BTSeq™ sequencing service (Celemics, Seoul, Republic of Korea). The positive lines were sequenced to identify any mutations (insertion, deletion, nucleotide base transition, or transversion) within the target regions. The nucleotide sequence data obtained were aligned and translated into amino acid sequences. Comparisons with the inbred line, 'CT001', reference of the BrLFY paralogs were performed using the CLC sequence viewer software 8.0 (Qiagen, Aarhus, Denmark).
The expression levels of BrLFY paralogs were compared between the inbred line 'CT001' and E 0 LFY-edited lines (LFY-7 and LFY-13) grown under long-day (16 h light/8 h dark) photoperiodic conditions. Total RNA was extracted from frozen tissue using the P&C Rapid RNA Prep Kit (Biosolution, Suwon, Republic of Korea). cDNA was synthesized from each sample using BioFACT™ RT-Kit (Biofact, Seoul, Republic of Korea). Each qPCR reaction contained 1 µL of cDNA, 2 µL of each primer, 5 µL of BioFACT™ 2X Real-Time PCR Master Mix (Biofact, Seoul, Republic of Korea), and water to a final volume of 10 µL. The qPCR cycling conditions began with a denaturing phase at 95 • C for 15 min; 35 cycles of 95 • C for 20 s, 55 • C for 20 s, and 72 • C for 20 s; and a final extension step at 72 • C for 5 min. The expression levels of BrLFY paralogs were normalized using the BrActin7 gene, and each reaction was performed with three biological replicates. The relative expression levels of E 0 LFY-edited lines were calculated using the 2 −∆∆Ct method.

Bolting Time Observation of the Inbred Line, 'CT001', and LFY-Edited Lines
To characterize the bolting time of the LFY-edited lines, 'CT001' and LFY-edited lines were artificially treated for 8 weeks in a cold room at 4 • C under 16 h/8 h (light/dark) photoperiod to induce a transition to the reproductive growth phase. The pots planted with 'CT001' and the LFY-edited lines were transferred to a greenhouse at 23 • C after cold-treatment. The status of bolting occurrence on a specific date was compared. After vernalization, the days of bolting and the first floral axis observations were recorded. The growth stages of the individual lines were observed at regular intervals. The selected LFY-edited lines were then advanced to the next generation.

Identification of the Number of T-DNA Copies and Insertion Position
Southern hybridization was performed to confirm the T-DNA inserted Chinese cabbage genome. A restriction enzyme with as many cleavage sites as possible in the inbred line 'CT001' genome was investigated. Genomic DNA was cut using the EcoRI restriction enzyme, which has approximately 45,000 cleavage sites in the genome. Although the EcoRI restriction enzyme has several cleavage sites in the genome, it is not present in the probe sequence. Therefore, normal hybridization was expected to occur. To predict the approximate molecular size of the product in which the hybridization signal occurred, lambda-HindIII molecular markers were loaded together. DNA fragments were separated overnight on a 1.0% agarose gel at 20 V and transferred to a Hybond N + nylon membrane (Amersham Pharmacia, Buckinghamshire, UK). The membrane was UV-crosslinked and hybridized with a hybromycin DNA fragment with α-32 P-labeled dCTP using BcaBEST labeling (TaKaRa, Otsu, Japan). The hybrid membrane was washed twice in a shaking incubator and exposed to an X-ray film in a cassette for 7 days.
To determine the T-DNA insertion site of the E 0 LFY-edited lines, an effective variable argument thermal asymmetric interlaced PCR (VA-TAIL PCR) analysis was used [48]. The developed Chinese cabbage-specific AD primers and T-DNA-border-specific amplifying primer sets were used to amplify the junction region of T-DNA introduced into the LFYedited line (Table S2 in Supplementary Materials). VA-TAIL PCR analysis was performed using 2X Taq PCR Master Mix (Biofact, Seoul, Republic of Korea) in a thermocycler (Applied Biosystems, Carlsbad, CA, USA). The total reaction volume was 30 µL and comprised 100 ng gDNA, AD, border-specific primers, and dH 2 O. The VA-TAIL PCR method and program conditions used in this study were the same as described in a previous study [49]. Specific amplicons were eluted using a P&C Multiple Elution Kit (Biosolution, Suwon, Republic of Korea), and the elution products were sequenced using the BTSeq TM Contiguous Sequencing Service (Celemics, Seoul, Republic of Korea). Finally, the T-DNA insertion position (intergenic, exon, intron, 5 upstream-1000, and 3 downstream-300) was analyzed using the obtained sequence of the junction region.

Conclusions
LEAFY (LFY) is a transcription factor that determines floral identity and acts as a flowering inducer by interacting with various flowering-related genes. The BrLFY gene was structuralized and sgRNA was designed for CRISPR/Cas9 gene-editing vector construction. Agrobacterium-mediated transformation of Chinese cabbage plants was performed. By analyzing the characteristics of the selected LFY-edited lines, the bolting time was found to be delayed, even after the vernalization. CRISPR/Cas9-mediated mutagenesis of BrLEAFY disrupted the normal formation of amino acids, which delayed the bolting and flowering of Chinese cabbage. The transgene-free LFY-edited lines displayed the inheritance of base mutations and late bolting. These findings demonstrate that the CRISPR/Cas9 technology is an effective tool for mutagenesis in Chinese cabbage and that the BrLFY gene mutation influences the bolting time. Overall, to our knowledge, this is the first study to achieve effective BrLFY gene editing using the CRISPR/Cas9 system.

Patents
A patent applicated in Korea (patent application number 10-2021-0143792; application date 21 September 2021) with the following titles; "Gene-editing vector for Brassica rapa plant with late flowering trait and transformation method using the same".