CRISPR/Cas9-mediated gene editing to confer turnip mosaic virus (TuMV) resistance in Chinese cabbage (Brassica rapa)

Abstract Genome editing approaches, particularly the CRISPR/Cas9 technology, are becoming state-of-the-art for trait development in numerous breeding programs. Significant advances in improving plant traits are enabled by this influential tool, especially for disease resistance, compared to traditional breeding. One of the potyviruses, the turnip mosaic virus (TuMV), is the most widespread and damaging virus that infects Brassica spp. worldwide. We generated the targeted mutation at the eIF(iso)4E gene in the TuMV-susceptible cultivar “Seoul” using CRISPR/Cas9 to develop TuMV-resistant Chinese cabbage. We detected several heritable indel mutations in the edited T0 plants and developed T1 through generational progression. It was indicated in the sequence analysis of the eIF(iso)4E-edited T1 plants that the mutations were transferred to succeeding generations. These edited T1 plants conferred resistance to TuMV. It was shown with ELISA analysis the lack of accumulation of viral particles. Furthermore, we found a strong negative correlation (r = −0.938) between TuMV resistance and the genome editing frequency of eIF(iso)4E. Consequently, it was revealed in this study that CRISPR/Cas9 technique can expedite the breeding process to improve traits in Chinese cabbage plants.


Introduction
Plant viruses pose a severe threat to global crop yield. There are many approaches to developing virus-resistant cultivars in crop plants. Among them, one common approach is traditional plant breeding by incorporating virus resistance genes from wild relatives into elite cultivars [1][2][3][4]. A recent alternative approach is genome editing, which instantly allows the importation of alleles mediating resistance into crop plants, preventing long and laborious backcrosses in traditional breeding [1,5]. The genomeedited crops can be transgenic free [5,6], and they probably would be classified as non-genetically modified food crops [7]. Thus, the genome editing technique may be a publicly acceptable advanced breeding method and opens up a new avenue for developing virusresistant cultivars.
CRISPR/Cas9 (clustered regularly interspersed palindromic repeats/CRISPR-associated protein 9) is a site-specific genome editing approach originating from the adaptive immune system of Streptococcus pyogenes against bacteriophages [8]. CRISPR/Cas9 technique was first reported in 2012, and ever since, it has turned out to be the most frequently used technology among plant biologists due to its low cost, simplicity, and considerably smaller timespan for construct formulation compared to alternative genome editing techniques [9]. Accurate genome editing of host factors can be used to develop recessive genetic resistance in plants against viral diseases [10]. Nevertheless, the implementation of CRISPR/Cas9 technology has not been frequently employed to deploy genetic resistance against pathogens, except for a several reports [1,5,[11][12][13][14][15].
Potyviridae, the most prominent family of plant RNA viruses, account for around 30% of reported plant viruses, causing many significant injuries to crop plants [16,17]. Especially, among the potyviruses, such as TuMV has been reported to severely threaten Chinese cabbage (Brassica rapa) crops [18]. Interestingly, plant RNA viruses entail host factors to sustain their life cycle [19,20]. Several genes conferring resistance to viruses are recessive (Kang et al. 2005a; [21]), including eukaryotic translation initiation factor (eIF) genes, for example, eukaryotic translation initiation factor 4E (eIF4E), eukaryotic translation initiation factor 4G (eIF4G), and eukaryotic translation initiation factor (iso)4E (eIF(iso)4E) [19,20]. Different plant species contain different numbers of eIF4E family members. In Arabidopsis, eIF(iso)4E is a significant host factor in TuMV resistance [22][23][24][25][26]. Following this, the Brassica eIF(iso)4E gene has been reported to be tightly associated with the Brassica recessive resistance genes retr02 and trs [27,28]. It was purposed by these results that the eIF(iso)4E gene could be a suitable target for the genome editing of TuMV resistance in Brassica species.
The CRISPR/Cas9 genome technology has been applied for developing potyvirus-resistant crops [1,5,12,14]. For example, the genome editing of the gene eIF(iso)4E in A. thaliana conferred resistance against the turnip mosaic virus (TuMV) [14]. It was revealed in another research that the CRISPR/Cas9-mediated genome editing of eIF4E in cucumber resulted in broad-spectrum viral resistance to numerous plant viruses, such as cucumber vein yellowing virus (CVYV), papaya ringspot mosaic virus-w (PRSV-W), and zucchini yellow mosaic virus (ZYMV) [1]. Likewise, resistance against the rice tungro spherical virus was conferred by a novel allelic mutation of rice eIF4G, developed using the CRISPR/Cas9 genome editing approach [12]. Recently, editing the eIF4E in tomatoes also resulted in resistant tomatoes against the pepper mottle virus (PepMoV) [5].
We developed TuMV-resistant Chinese cabbage (B. rapa) by mutating eIF(iso)4E in the Chinese cabbage cultivar "Seoul" using the CRISPR/Cas9 approach and Agrobacterium-mediated transformation in this study. We generated the T 0 transgenic plants carrying eIF(iso)4E mutations and produced T 1 progeny. We confirmed the indel frequency within the target regions of the edited mutant T 0 and T 1 plants and examined the resistance of T 1 plants to TuMV. Overall, this study provides a protocol for analyzing and interpreting Chinese cabbage CRISPR/Cas9 mutants and improving the targeted traits in crops using genome editing technology.

Transformation and generation of Chinese cabbage
We inserted the CRISPR/Cas9 constructs (pECO101-hyg-Cas9-eIF(iso)4E vector) carrying the corresponding sgRNAs into the B. rapa cultivar, "Seoul," through Agrobacterium-mediated transformation to develop Chinese cabbage plants in which the eIF(iso)4E gene has been edited (Fig. S1). We cultured the shoot in a shooting medium to increase the elongation of the shoot induced by the callus. The regenerated shoot was relocated to the rooting medium to promote root formation (Fig. S1). PCR analysis was performed using Cas9-specific primers to confirm and select a transgenic entity harboring transfer DNA (T-DNA) in a regenerated plant ( Table 2). Thirteen of the 15 regenerated plants confirmed the Cas9 transgenes by showing PCR products with the expected amplicon sizes (600 bp). The transformation efficiency was 86.7% ( Fig. 2A).

Confirmation of CRISPR/Cas9-induced T 0 plant to gene editing efficiency
We performed the targeted deep-sequencing using the PCR product to ensure whether the CRISPR/Cas9 induced mutations in the eIF(iso)4E gene. Unfortunately, we confirmed that only one sgRNA out of three sgRNAs constructed into the pECO101-hyg-Cas9-eIF(iso)4E vector sgRNA3 targeting three eIF(iso)4E genes had high editing efficiency. Among the eIF(iso)4E-related genes, we was confirmed for gene correction only for Bra035393, a TuMV recessive resistance-related gene reported by Kim et al. [27]. The T 0 edited plants #4, #6, #7, and #12 were confirmed by deep sequence analysis that had high gene editing efficiency among the selected plants. Four types of sequence variations were observed: one nucleotide (A, T, or C) insertion (+1) into the target region or twenty eight nucleotide deletion (−28), including the protospacer adjacent motif (PAM) sequence (Fig. 2B). The total indel frequencies were 47.4% for #4, 47.2% for #6, 62.8% for #7, and 49.1% for #12 T 0 edited plants. The primary indel frequencies were 21.1%, 43.6%, 52.8%, and 20.2%, respectively (Fig. 2B). No homozygous mutant lines were found among the T 0 edited plants.

Identification of Indel pattern and frequency in T 1 edited plants
For propagation to T 1 seeds, #6 and #7 T 0 edited plants, which consisted of one significant indel pattern among the four T 0 edited plants (#4, #6, #7, and #12), were self-fertilized (Fig. 2B). We extracted gDNA from the leaf samples of the #6 and #7 T 1 edited plants and verified the existence of transgene by PCR with transgene-specific primers ( Table 2). Deep sequence analysis was performed on the T 1 -edited plants to measure the editing efficacy ( Fig. S2A and B). Although the T 0 -edited plants had one significant indel pattern, and the generational progression was achieved using self-fertilization, the T 1 -edited plants showed various new indel patterns. We divided the different indel patterns into three groups: (i) single, (ii) double, and (iii) mosaic (Fig. 3). In the group having a single indel pattern, #6 and #7 T 1 edited plants showed the same indel pattern as T 0 edited plants. In addition, #7-53, #7-62, 7-66, and #7-72 showed another nucleotide (A) insertions, and #7-79 showed PAM site deletions (Fig. 3A). In the group having a double indel pattern, one of the two indel patterns showed the same pattern as T 0 plants. In the case of #6-41 and #6-42, low mutant frequency, but a different nucleotide (A) was inserted. Plant #6-79 had two deletions that were not observed in T 0 (Fig. 3B). In #7-73 and 7-78, PAM sites were deleted, and in #7-74 and #7-84, different nucleotides (A) were inserted (Fig. 3B). In the group having mosaic indel pattern, a nucleotide different from the significant pattern of T 0 was inserted in most T 1 edited plants (Fig. 3C). The highest total indel frequency was #6-62, #6-64, and #7-79, and were 99.8%, 99.8%, and 100, respectively (Fig. 3). The lowest total indel frequency was 55.0% for #6-76 and 6.6% for #7-81.

TuMV resistance in eIF(iso)4E T 1 edited plants
We inoculated TuMV into the single and double indel pattern T 1 edited #6 and #7 T 0 plants and wild-type control (Fig S3A and B) to examine whether these CRISPR/Cas9-derived eIF(iso)4E mutants of Chinese cabbage are resistant to TuMV. TuMV caused several leaf symptoms in infested plants, for example, conventional systemic mosaic, leaf curling, chlorotic lesions, chlorotic mottling, vein clearing, necrotic lesions, and stunted plant growth. The wild-type plant displayed distinctive TuMV symptoms as early as seven days post inoculation (DPI), containing vein clearing and several small chlorotic mottling in the leaves (Fig. S3C). We performed the DAS-ELISA analysis using the inoculated leaves of #6 and #7 T 1 edited plants after 21 DPI (Fig. 4) to verify virus infection. We detected considerable amounts of virus coat protein accumulation in the leaves of the susceptible wild-type control. However, coat protein hardly accumulated in the systemic leaves of any edited mutant plants, validating that the eIF(iso)4E-edited mutant plants were resistant to TuMV (Fig. 4).

Differences in TuMV resistance according to gene editing efficiency
Next, we measured the correlation between the total indel frequency of the T 1 edited plants and ELISA scores to clarify the role of eIF(iso)4E in TuMV resistance. The #6-57 plant, which showed the highest susceptibility when observed at seven DPI (not shown data), had 8.5% of the total indel frequency and 2.0485 of the ELISA score at 21 DPI as a susceptible plant. The #6-62 and #6-76 edited plants with 99.8% and 55.0% of total indel frequency showed 0.1094 and 0.7593 ELISA scores, respectively. It was revealed in the correlation analysis that the total indel frequency had a high negative correlation (r = −0.938) with the ELISA score (Fig. 5),  confirming that eIF(iso)4E was a significant host factor in TuMV resistance.

Phenotype evaluation of TuMV resistance edited plants
We confirmed TuMV symptoms between edited and wild type (WT) plants 30 DAI (days after inoculation) to evaluate the phenotype (Fig. 6A). Edited plants that showed resistance to TuMV and small symptoms of TuMV by the total indel frequency, but there were no experienced any major issues with growth and development (Fig. 6B). While, WT plants exhibited mosaic symptoms, which are typical of TuMV, throughout their leaves, resulting in poor conditions for growth and development (Fig. 6B). We transplanted the inoculated plants into pots and grew them with the non-inoculated plant (mock) in an LMO glass house to evaluate the resistance of the wild type (WT) and edited plants to TuMV. After 30 days of planting in a pots, phenotypic evaluation was performed with the naked eye. Likewise, all the edited plants were resistant to TuMV, the phenotypic difference was observed according to the total indel frequency. Plant #6-62 with a high total indel frequency (99.8%) showed no phenotypic difference from the mock, but in the case of plant #6-76 with a medium total indel frequency (55.0%), there was no problem in growth. Still, mosaic symptoms, one of the viral symptoms, appeared in the leaves. Also, mosaic symptoms appeared in newly emerging shoots (Fig. 7). Most of the WT plants died, and even if they survived as mature plants, mosaic symptoms were severe and there was a noticeable problem in growth and development (Fig. 7).

Discussion
Food security is one of the major problems confronting the modern world due to the constantly increasing global population. According to estimates, global food production should be doubled in the next few decades to cope with ever-increasing consumer requirements [14,29]. It has been appraised that viral infection in Table 2. List of primers used in the current study.

Name Primer sequence (5 to 3 ) Amplicon size (bp) Purpose
Hyg_F GCGAAGAATCTCGTGCTTTC  209  Mutation detection  Hyg_R  CAACGTGACACCCTGTGAAC  MY5 -AtU6 _F  AAGAAGAGAAGCAGGCCC  600 Transgene confirmation YM20 R -eIF4E-gRNA3_R AAACCCGCCGCAAGGGCTTCGTTC food crops ensues around 10-15% yield loss annually worldwide (Kang et al. 2005a; [30]). Extenuation of these losses by incorporating viral resistance into existing elite verities is a promising approach to achieving global food production goals. Exploiting genetic resistance in crops is the most endurable strategy to manage virus infection (Kang et al. 2005a; [31,32] [2]). Furthermore, transgenic methods to introgress resistance in various crops have been effectively used against several viruses in the last couple of decades [33]. Nonetheless, complications encountered in regulatory provisions and socio-economic issues, such as the public unacceptability of transgenic crops, are significant constraints to widely implementing these biotechnological approaches. In contrast, genome-edited crops can be transgenic free [5,6,34]. They would probably be classified as non-genetically modified food crops [7].
Since recessive resistance genes are deemed to provide more durable resistance compared with dominant resistance genes in the present study, we focused on developing CRISPR/Cas9edited mutants to establish virus resistance in Chinese cabbage. We targeted the eIF(iso)4E gene using the CRISPR/Cas9 system to achieve this, and developed TuMV resistance to potyvirus in B. rapa Chinese cabbage, cultivar "Seoul," through Agrobacteriummediated transformation.
In Arabidopsis, mutations in eIF(iso)4E resulted in TuMV resistance, and there is only a single copy gene [24]. In contrast, B. rapa has multiple copies of the eIF(iso)4E gene [27,35], and each gene was sequenced using cloned genomic DNA by Kim et al. [27]. We found three copies related to eIF(iso)4E in B. rapa and confirmed the sequence in NCBI to distinguish the gene for each copy (Table 1). In several previous studies [27,28], the TuMV-related recessive gene was reported as Bra035393 (BraA.eIF(iso)4E.a), whereas Bra035531 (BraA.eIF(iso)4E.c), it has been reported that there is no specific variation between susceptible and resistant sequences [27]. Therefore, we first performed the gene editing analysis focusing on Bra035393 gene, and as a result of this study, edited-plant with resistance to TuMV obtained. Since we targeted TuMV resistance, we performed gene editing analysis targeting only Bra035393 by referring to the results of previous studies in this study, but we plan to additionally perform gene editing analysis for the other two copies of the gene (Bra035531 and Bra039484).
The eIF(iso)4E edited mutant plants showed indels eventuated at numerous locations within the sgRNA target site and, in some cases, near the sgRNA target region (Fig. 2 and 3). In our results, the eIF(iso)4E-targeted mutant plants in the T 0 generation were double indel patterns. However, we observed a mixture of mutant variations, including single, double, and mosaic indel patterns in T 1 -edited plants. The mosaicism has been reported previously in various edited plants [1,5,14,34,[36][37][38][39][40]. The mosaic indel patterns of the T 1 edited plants might be because the Cas9 protein and sgRNA are highly expressed in some edited plants. Moreover, additional alleles produced from lately-emerging chimeric tissues could also be unnoticed in T 0 plants, consequential ensuing in different f lowers exhibiting different alleles [36]. Because of this complexity, we advanced the T 0 plants with a simple primary indel pattern for the gene of interest in the next generations (Fig. 2). It was confirmed in further sequence analysis that the indel frequency in T 1 edited plants (Fig. 3), suggesting that their mutant status was stably transmitted to their progenies as generation progressed [1,5].
Kim et al. [18] have developed transgenic Chinese cabbage by targeting eIF(iso)4E at the protein level to develop broadspectrum resistance against potyviruses, such as TuMV (Turnip mosaic virus). TuMV particle accumulation was abridged when eIF(iso)4E proteins mutated in these amino acids were over-expressed in transgenic plants. Furthermore, reduced reciprocity between eIF(iso)4E and TuMV VPg was shown by the results to potentially cause conferred resistance [18]. However, the resistance mechanism remained unclear and required further studies to mutate eIF4E in Chinese cabbage and to study the variations at the DNA sequence level.
Consequently, we evaluated our eIF(iso)4E edited T 1 plants for TuMV resistance to the potyvirus (Fig. 4). Genome-edited plants harboring mutations in eIF(iso)4E conferred resistance against TuMV. We showed that there was a strong negative correlation (r = −0.938) between gene editing efficiency related to indel frequency and TuMV resistance evaluation using the ELISA score (Fig. 5). Additionally, we confirmed that homozygous lines show uniform resistance to TuMV resistance in an LMO glass house (Fig. 6). While no significant phenotypic difference was observed between the eIF(iso)4E-edited plants and wild type plants under our growth experimental conditions, it is possible that the mutation in the eIF(iso)4E gene could have adverse effect on growth of Chinese cabbage under different stress conditions. Recently, genome editing in Chinese cabbage has been performed for different traits, such as FLOWERING LOCUS C (FLC) for early f lowering, BrVRN1 gene for delayed f lowering, and PR55/B gene for self-incompatibility [41][42][43]. Moreover, DNA-free (transgene-free) genome editing by targeting the vernalization determinant FRIGIDA and phytoene desaturase gene (FRI and PDS) genes has also been done to pave toward minimizing GMO legislation-related issues [34]. However, no CRISPR/Cas9based genome editing study has been conducted to achieve viral resistance in Chinese cabbage. In the current study, we applied CRISPR/Cas9 gene editing to induce mutations in the Chinese cabbage eIF(iso)4E gene by Agrobacterium-mediated transformation. Different indel patterns of mutations in the eIF(iso)4E gene, including mosaic, double, and single, were detected and verified by sequencing in T 0 and subsequent T 1 generations. Our work demonstrated effective CRISPR/Cas9 editing of Chinese cabbage eIF4E1, triggering resistance to potyvirus. The findings of the current study can be practically extended to the breeding programs of relative Brassica spp. and will help develop virusresistant cultivars. Hence, targeted genome editing could be anticipated to expedite plant breeding for disease resistance, particularly for virus resistance in crop plants, by enabling the incorporation of precise and expected genetic modifications directly into an elite cultivars background.

Plant materials and conditions for growth in vitro
The Chinese cabbage cultivar, "Seoul" (Dong-bu Seed, Korea), which has been reported to be susceptible to turnip mosaic virus [18], was used for Agrobacterium-mediated transformation. The seeds used in this study were first disinfected through seed coat sterilization in 70% ethanol for 1 to 2 min, followed by shaking in 10% commercial Clorox for 10 to 15 min at 100 rpm. The seeds were thoroughly rinsed five to six times with sterilized water. The sterilized seeds were germinated on 1/2 MS medium [44] supplemented with 2% sucrose and 0.8% agar. These seeds were subjected to in vitro cultivation to grow hypocotyls up to 6-8 cm for five days at 25 • C. The hypocotyls were cut, avoiding the apical meristem and quickly cut into 7-8 mm segment intervals before  the first true leave emerged. Forty explants were cultured in a 15 x 90 mm petri dish containing 0.8% MS medium (pH 5.8) supplemented with 3% sucrose, BA (4 mg/ml), NAA (1 mg/ml). All of the cultures were incubated in the controlled incubation room for 3 days at 25 • C under dark condition. These explants were used for tissue culture and Agrobacterium-mediated transformation.

Construction of single guide RNAs (sgRNAs) and CRISPR/Cas9 vector
The sequences of eIF(iso)4E, which have been the Chinese cabbage potyvirus, as previously reported by Jenner et al. [35], were downloaded from the NCBI (http://www.ncbi.nlm.nih.gov/) ( Table 1). The BLAST searches were performed using the Brassica database (BRAD, http://brassicadb.cn/) to obtain sequences around eIF(iso)4E in B. rapa. The eIF(iso)4E genes (Bra035531, Bra039484, and Bra035393) of Chinese cabbage were aligned and selected as the target sequences (Fig. 1A). The CRISPR RGEN tools (http://www.rgeneome.net) was used to identify appropriate target sgRNA sequences and designed sgRNAs common for each Chinese cabbage eIF(iso)4E gene (Fig. 1B). All three sgRNAs (sgRNA1, 2, and 3) were designed to target Bra035531, Bra035393 and Bra039484 genes (Fig. 1A). To express Cas9 protein in plants, the maize codon-optimized Cas9 driven by the CaMV 35S promoter. The pECO101 vector with the Cas9 gene was cloned into a binary vector under the control of the AtU6 promoter (Fig 1B). The sgRNAs scaffold was annealed and cloned into a BsaI restriction site in the vector pECO101 by Golden-Gate cloning according to a previously reported method [45]. The resulting CRISPR/Cas9 vectors were transformed into Agrobacterium tumefaciens strain GV3101 by heat shock.

Agrobacterium-mediated transformation of Chinese cabbage
A single colony of Agrobacterium was grown in 10 mL of liquid YEP medium having 50 mg/mL spectinomycin at 28 • C until it reached an OD 600 value of 0.6-0.8. The pellet was harvested by centrifuge at 13000 rpm for 15 min at 4 • C using Agrobacterium suspension.. The gained pellet was re-suspended entirely in liquid 1/2 MS medium (pH 5.2) containing 3.6% glucose. The Agrobacterium resuspension was applied to infect hypocotyl explants and cocultured for 1 h, after infection, they were followed by transfer to sterile filter paper to clear remained suspension completely. For co-culture with agrobacterium, these explants were placed on the MS medium containing 3% sucrose, BA (4 mg/ml), NAA (1 mg/ml), and 0.8% plant agar (pH 5.2) and incubated in two days at 22 • C and dark condition. The co-cultured explants were transferred to the MS selection medium containing 3% sucrose, AgNO 3 (4 mg/L), IBA (4 mg/L), NAA (3 mg/L), hygromycin (10 mg/L), and 0.8% plant agar (pH 5.6). The number of calli and shoots formed on the explants was determined after cultivation for three and 10 weeks, respectively. Finally, the regenerated shoots were transferred to a rooting medium containing 1/2 MS medium, 3% sucrose, and 0.5% plant agar (pH 5.8). The regenerated plants were cultivated in a growth room maintained at 22 • C and a 16 h light/8 h dark cycle.

DNA extraction and PCR analysis
Total genomic DNA (gDNA) was extracted from the leaf samples of potential transgenic plants using the cetyltrimethylammonium bromide (CTAB) method [46]. gDNA was enumerated using a Nanodrop spectrophotometer (Nanodrop Technology, Inc., Wilmington, DE, USA) and diluted to 50 ng/mL. Transgenic plants were evaluated by PCR analysis using HPT (Hygromycin phosphotransferase) and Cas9 gene-specific primers ( Table 1). The amplification settings was adjusted as follows: pre-denaturation at 95 • C for 10 min, followed by 35 cycles of 95 • C for 30 s, 58 • C for 30 s, 72 • C 30 s, and final extension at 72 • C for 10 min.

Target deep-sequencing and mutation detection
Plant that are transgenic were assessed for mutation identification using specific target primers linked to sgRNA target regions ( Table 1). The samples detected with target site insertion were subjected to deep sequence analysis. For the deep-sequencing analysis, three rounds of PCR were performed, and during the first PCR, analysis primers were designed with a amplication size of 600-800 bp. PCR was performed with a final volume of 20 μL, containing 50 ng template DNA concentration, 1.0 pmol of reverse and forward primers, 10 mM dNTPs, 10x Hipi buffer, and 0.2 units of Hipi plus Taq polymerase. The following protocol was used for PCR amplification: 95 • C preheating for 3 min and 95 • C for 30 s, followed by 10 cycles of 72 • C for 30 s, 72 • C for 45 s, and 20-30 cycles at 95 • C for 30 s, 62 • C for 30 s, 72 • C for 45 s, and a final extension of 5 min at 72 • C. The second round of PCR was used to attach the adapter, and the primers were designed with a product size of 150-250 bp. The cleaned PCR products were subjected to index amplification with known barcodes. Next, index PCR was performed using a dual indexing and adapter kit (Illumina, California, USA). Finally, the PCR products were purified using a PCR clean-up kit (Cosmo Genetech, Seoul, Korea). The refined PCR products were sequenced directly according to the method by Shokralla et al. [47].

Inoculations of the TuMV virus
Frozen TuMV stock stored at −80 • C was used to prepare the virus inoculum. A preliminary experiment was conducted by inoculating 3-week-old plants of susceptible Chinese cabbage to confirm the virulence of the strain. Brief ly, frozen inoculate was crushed in 0.1 M potassium phosphate buffer (pH 7.0), combined with 400-grit carborundum, and scrubbed on the lower leaves of susceptible plants. Subsequently, the leaves were washed with distilled water after 10-20 min of inoculation [48]. The symptomatic leaves showing typical TuMV symptoms were harvested four weeks after inoculation and double checked with a doubleantibody sandwich enzyme-linked immunosorbent assay (DAS-ELISA) test kit (Agdia Inc., Elkhart, IN, USA) for the virus presence. The putative genome-edited plants and susceptible controls were inoculated, as described above. In detail, Chinese cabbage plants with 2-4 true leaves were used for viral inoculation and each plant one pair cotyledons were inoculated. The inoculated and noninoculated control plants (mock) were grown in a growth chamber at 26 • C under 16 h/8 h light/dark conditions in white f luorescent light.

Evaluation of resistance to turnip mosaic viruses
The inoculated plants were regularly inspected for the appearance of symptoms after viral inoculation. Leaves were observed developing typical TuMV symptoms [18] to discriminate resistant and susceptible plants after seven days of viral inoculation. The leaf tissues were subjected to DAS-ELISA to evaluate the presence of the virus. TuMV viral accumulation was finally assessed 21 days after inoculation. DAS-ELISA was conducted to identify the accretion of the coat protein of TuMV. Three replicates of inoculated and upper non-inoculated leaves of T 1 lines were assessed for the ELISA test. The absorbance value of each sample was measured at 405 nm using an ELISA reader (Titertek, Huntsville, AI, USA). The statistical significance of the data was determined with a t-test using the R Studio v. 4.0.3 package.