<i>Bamboo mosaic virus</i>‐mediated transgene‐free genome editing in bamboo

Abstract

The common method of delivering CRISPR/Cas reagents for genome editing in plants involves Agrobacterium-mediated transformation or preassembled CRISPR/Cas9 ribonucleoprotein complex delivery (Woo et al., 2015; Toda et al., 2019; Ye et al., 2020). These methods require labor-intensive and time-consuming plant tissue culture processes (Huang et al., 2022). Unfortunately, most plants exhibit extremely low efficiency in callus induction and regeneration; these technical challenges greatly hinder the application of genome editing. Recent developments in plant RNA virus-based expression vectors (Ma et al., 2020; Chen et al., 2022) provide a convenient, efficient, and cost-effective way for DNA-free genome editing in plants, leveraging the fact that virus RNA does not integrate into the genome. However, the stability of virus vectors is negatively correlated with the size of the inserted foreign genes. Consequently, achieving efficient expression of Streptococcus pyogenes Cas9 (SpCas9, c. 4.2 kb) by virus-based vectors remains challenging. Most reported viruses capable of delivering Cas9 proteins are negative-strand RNA viruses (Ma et al., 2020; Liu et al., 2023; Zhao et al., 2024), with only a few positive-strand RNA viruses identified (Uranga et al., 2021; Lee et al., 2024). Thus, delivering virus-based sgRNA vectors to plants overexpressing Cas9 is the most commonly used strategy (Ali et al., 2015; Jiang et al., 2019; Li et al., 2021). However, it is difficult to use this method to generate a Cas9-free mutant by crossing with wild-type (WT) plants with long flowering cycles, such as bamboo (Ye et al., 2017). Bamboo mosaic virus (BaMV) has a typical flexible filamentous virion structure with the positive-sense single-stranded RNA genome (Hsu et al., 2018). The BaMV-mediated expression system can effectively drive the expression of large foreign gene fragments (Jin et al., 2023). For the first time, we developed a BAMV-mediated Cas protein and sgRNA delivery system in WT Nicotiana benthamiana and bamboo. This approach enables targeted gene editing in noninfected leaves or stems in bamboo without the need for Cas9-expressing transgenic lines, leveraging BaMV's large cargo ability to transport Cas9 proteins. Nicotiana benthamiana seeds were germinated on 1/2 MS medium (M519; Phytotech, Lenexa, KS, USA), supplemented with 30 g l−1 sucrose, and cultured at 28°C in a growth chamber with a 16 h : 8 h, light : dark photoperiod. The green seedlings were then transferred to soil and cultivated in a glasshouse maintained at 23–25°C with 60% relative humidity and a 15 h : 9 h, light : dark photoperiod. For Dendrocalamus latiflorus Munro and Phyllostachys edulis, peeled seeds were soaked in water for 24 h before being evenly sown uniformly on nutrient-enriched soil and covered with 1-cm thick nutrient soil. These seeds were germinated and cultivated in a glasshouse at 23–25°C under a 16 h : 8 h, light : dark photoperiod. To construct BaMV vectors for expressing Cas9, two variants of AsCas12f1 (Ye et al., 2024), and sgRNA, we modified the previously developed BaMV-Cas9 expression vector (Jin et al., 2023). This involved the addition of the coat protein (CP) promoter of BaMV and an Xba I restriction site following the termination codon of the BaMV triple gene block protein 3 (TGBp3). We then inserted an 83 bp scaffold RNA (for Cas9) or 94 bp scaffold RNA (for HKRA and YHAM) and guide RNA sequences specific to N. benthamiana (gNbPDS), P. edulis (gPheRDR6) or D. latiflorus (gDlmRDR6) (Supporting Information Table S1; Fig. S1a) at the Xba I site, under the control of the CP promoter. The constructed BaMV vector, containing SpCas9 and sgRNA, was designated as pBaMV-Cas9, and containing two variants of AsCas12f1 and sgRNA. BaMV vectors were designated as pBaMV-HKRA and pBaMV-YHAM (Figs 1a, S2). Previous studies have shown that synthetic constructs containing tandemly arrayed tRNA-gRNA structures are capable of precise processing into gRNAs to achieve simultaneous targeting of multiple genetic loci (Xie et al., 2015). Additionally, we also designed the pBaMV-g1tg2t-Cas9 vector for multiplexed genome editing by attaching a 77-bp tRNAGly sequence (Ellison et al., 2020) to the 3′ end of the sgRNA scaffold sequence and concatenated two sgRNAs into an sgRNA1-tRNAGly-sgRNA2-tRNAGly (g1tg2t) structure (Fig. 1a; Table S1). To construct the pBaMV-nCas9 and pBaMV-cCas9 vectors, we split Cas9 into 573-aa nCas9 and 794-aa cCas9 (Truong et al., 2015), and then fused the split Npu_DnaE (nIntein and cIntein) with split-Cas9. The Split-Cas9 protein after fusion was separately assembled into the BaMV subgenome (Fig. S2). The constructed pBaMV-Cas9 or pBaMV-AsCas12f1 plasmids were transferred into Agrobacterium tumefaciens strain GV3101 using the freeze–thaw method. The plasmid was cultured on Yeast Extract Mannitol Broth (YEB) solid medium containing 50 mg l−1 kanamycin and 25 mg l−1 rifampicin for c. 48 h. A single colony was selected and then inoculated in a liquid medium containing the same concentration of antibiotics. The plasmid was shaken at 28°C and incubated at 200 rpm until an OD600 0.5–0.8 was reached. Enriching bacteria was centrifuged briefly and washed once with tobacco infection buffer (10 mM MES, 10 mM MgCl2, 100 μM acetylsyringane, pH 5.6). Finally, Agrobacterium tumefaciens was resuspended in the tobacco infection buffer and incubated at 25°C in darkness for 2 h before inoculating tobacco leaves. After 10 d of infiltration, the infected leaves were ground with virus infection buffer (1% PVP, 50 mM tripotassium orthophosphate, 1‰ 2-Hydroxy-1-ethanethiol, pH 8.0) and diamond sand. Then, the juice obtained from grinding was used for the mechanical inoculation of bamboo leaves. After 2 d of moisturizing cultivation, the bamboo seedlings infected with BaMV were returned to normal growth conditions in a glasshouse at 23–25°C. Total RNA was extracted by TRIZOL method from 0.1 g fresh leaf or stem tissue of tobacco and bamboo using the spectrophotometer (DS-11 FX+; DeNovix, Wilmington, DE, USA) for quantitative assessment. According to the protocol of HiScript II 1st Strand cDNA Synthesis Kit (R211-01; Vazyme, Nanjing, China), first-stranded cDNA was synthesized from 2 μg of total RNA treated with DNase, using oligo-dT primers and BaMV 3′-UTR specific primers (Ba32). All primer sequences used for reverse transcription polymerase chain reaction (RT-PCR) analysis are listed in Table S1. According to the protocol provided, DNA extraction was performed from 0.15 g tobacco or bamboo leaves or stem following HiPure HP Plant DNA Kit protocol (D3187; Magen, Beijing, China). For PCR amplification of sgRNA target genes in these samples, high-fidelity enzyme 2 × Phanta Flash Master Mix (Dye Plus) (P520; Vazyme) was used. All primer sequences are listed in the Table S1. After purification, the obtained PCR products were subjected to restrictive digestion using Nco I (FD0574; Thermo Scientific, Colorado Springs, CO, USA) or T7 Endonuclease I (EN303; Vazyme). The resulting digested fragments were then detected via electrophoresis on 2% agarose gel. For Sanger sequencing, 200 ng of PCR products were digested with corresponding restriction enzymes, followed by TOPO-cloning into the easy vector using the Universal Zero TOPO TA/Blunt Cloning Kit (10906ES20; Yeasen, Shanghai, China). Six to 10 positive clones were sequenced. For amplicon deep sequencing, we designed Hi-TOM sequencing primers for the target genes according to the Hi-TOM Rapid Sequencing Service Protocol (http://www.hi-tom.net/hi-tom/) (Sun et al., 2024). The cloned amplicon sequences were sequenced, and the clean sequences were aligned to reference sequence using the Hi-Tom online platform. The editing efficiencies were calculated by dividing the number of reads with indels by the total number of sequenced reads. For protein extraction, samples from BaMV-infected tobacco and bamboo leaves were quickly frozen in liquid nitrogen and ground into powder. Subsequently, 300 μl of protein loading buffer (25 mM Tris–HCl pH 6.8, 0.5 M DTT, 10% (w/v) SDS, 0.5% (w/v) bromophenol blue, and 50% (v/v) glycerol) was added to each sample, and the mixture was homogenized and heated at 100°C for 10 min. Then, the samples were centrifuged at room temperature at 12 396 × g for 10 min. Proteins were separated using 12.5% SDS-polyacrylamide gel electrophoresis and subsequently transferred onto nitrocellulose (NC) membrane at 100 V for 90 min in 1× transfer buffer (25 mM Tris base, 192 mM glycine, 3.5 mM SDS, and 20% (v/v) solvent). Following the transfer, the NC membrane was washed with Tris buffer salt solution (TBS, 20 mM Tris base, 150 mM NaCl, pH 7.5) once, and then seal the membrane with TBST (TBS and 0.05% (v/v) Tween 20) containing 5% (w/v) skim milk powder for 1 h at room temperature with gentle agitation 50 rpm. The closed NC membrane was washed once with TBS and incubated with the primary antibody in TBST for 1 h at room temperature. After three washes in TBST (10 min each), the membrane was incubated with the secondary antibodies containing TBST for 1 h at room temperature. The membrane underwent three additional washes in TBST (10 min each), and a final rinse in TBS. Finally, the membrane was incubated with a chemiluminescent substrate in the dark on for 1 min and subsequently analyzed using the Amersham Imager 600. Ordinary one-way ANOVA and two-way ANOVA tests were performed with GraphPad prism v.10.2.0 (GraphPad Software Inc., San Diego, CA, USA) for the comparative analysis of editing efficiencies across different groups. To address the technical limitations of low-genetic transformation efficiency during bamboo callus regeneration, we engineered the pBaMV-Cas9 vector, integrating SpCas9 and sgRNA between the TGBps and CP of BaMV (Fig. 1a). Heterologous RNA expression is controlled by the CP promoter. To further enhance BaMV infectivity and increase expression levels of Cas9, we incorporated p19 from Tombusvirus, linked to SpCas9 via a 2A peptide. Subsequently, to assess the feasibility of the BaMV-based delivery system for targeted mutagenesis of endogenous loci in N. benthamiana, we selected the Phytoene Desaturase (PDS) gene for editing (Table S1; Fig. S1a). Four-week-old tobacco leaves were infiltrated with Agrobacterium tumefaciens carrying the BaMV-Cas9 construct targeting NbPDS (Fig. 1b). Approximately 3 wk after infiltration, we examined the fifth and sixth symptomatic systemic leaves (showing curling and chlorosis) at the molecular level to measure the expression levels of viral RNA, Cas9 and sgRNA transcripts, and Cas proteins (Fig. 1c,d). The PCR-RE assays utilizing Nco I endonuclease on DNA samples extracted from systemic leaves revealed mutagenesis frequencies ranging from 19% to 37% for NbPDS-A and 29% to 52% for NbPDS-B (Figs 1e, S1b). Additionally, amplicon deep sequencing of truncated NbPDS showed that the editing efficiency mediated by pBaMV-Cas9 ranged from 35.8% to 47.4%, with an average of 41.3% (Fig. 1f). Finally, Sanger sequencing of NbPDS-A and NbPDS-B further confirmed the various types of sequence editing at the targeted site (Figs 1g, S1c; Notes S1). After 30 d postinoculation (dpi) with pBaMV-Cas9, we also performed PCR-RE detection on the cDNA of NbPDS in symptomatic systemic leaves, which showed consistent mutations in NbPDS genome DNA as well (Fig. S1d). To evaluate the feasibility of multiplex mutagenesis using the BaMV-mediated CRISPR/Cas9 system, we constructed two pBaMV-g1tg2t-Cas9 vectors targeting either two distinct sites within the NbPDS gene (gNbPDS&gNbPDS-2) or both the N. benthamiana PDS gene and the RNA-dependent RNA Polymerase 6 (RNR6) gene (gNbPDS&gNbRDR6) (Fig. S1a). Two identical 77-base pair (bp) tRNAGly sequences were appended to the 3′ end of sgRNA, and two sgRNAs were concatenated to form the sgRNA1-tRNAGly-sgRNA2-tRNAGly (g1tg2t) structure (Fig. 1a; Table S1). Following infection with two pBaMV-g1tg2t-Cas9, we collected systemic leaves for PCR-RE detection, and the results revealed that both vectors could induce effective mutations at target sites despite the reduced efficiency compared to pBaMV-Cas9. The gNbPDS&gNbPDS-2 vector-mediated mutation frequencies of NbPDS-A and NbPDS-B ranged from 12% to 15% and 14% to 23%, respectively (Figs 1h, S1e). Amplicon deep sequencing of truncated NbPDS showed the editing efficiency of NbPDS-A by gNbPDS&gNbPDS-2 vector-mediated ranged from 4.14% to 5.79% (average 4.93%), similarly, the editing efficiency of NbPDS-B ranged from 5.49% to 8.72% (average 7.41%) (Fig. 1i). Sanger sequencing further confirmed that gNbPDS&gNbPDS-2 vector successfully caused sequence changes at two target sites in NbPDS-A and NbPDS-B, along with large fragment deletions between the two cleavage sites (Fig. 1j; Notes S1). The mutation frequencies based on PCR-RE assays for NbPDS-B and NbRDR6-B mediated by the gNbPDS&gNbRDR6 vector ranged from 21% to 34% and 21% to 26%, respectively (Figs 2a, S1f). Additionally, editing efficiency based on amplicon deep sequencing for NbPDS was between 7.39% and 16.88% (average 12.27%), while for NbRDR6, ranged from 12.09% to 16.67% (average 14.26%) (Fig. 2b). Simultaneously, Sanger sequencing also confirmed effective editing at both NbPDS and NbRDR6 target sites using the gNbPDS&gNbRDR6 vector (Figs 2c, S1g; Notes S1). To further substantiate the efficacy of the BaMV-based CRISPR/Cas9 system for genome editing in D. latiflorus and P. edulis, we targeted DlmRDR6 and PheRDR6 gene loci, respectively. After 30 dpi, the newly grown leaves exhibited typical mosaic symptoms (Fig. 1b), and immunoblotting confirmed the expression of Cas9 proteins in the new leaves of bamboo (Fig. S2a). Simultaneously, the expression levels of viral RNA, Cas9, and sgRNA transcripts in the stems were assessed using RT-PCR (Fig. S2b). Amplicon digestion of DlmRDR6 and PheRDR6 with T7 endonuclease I (T7EI) from symptomatic leaves and stems indicated successful target gene editing. In D. latiflorus, mutagenesis frequencies ranging from 9% to 19% in leaves and 6% to 16% in stems (Figs 2d, S2c). Similarly, in P. edulis, mutagenesis frequencies ranged from 5% to 16% in leaves and 4% to 16% in stems (Figs 2e, S2d). To validate the T7EI results, we conducted deep sequencing analysis of one 520-bp PCR product of DlmRDR6 and a 538-bp PCR product of PheRDR6. This amplicon deep sequencing showed that the pBaMV-Cas9 mediated an editing efficiency for DlmRDR6 ranging from 2.01% to 5.81% (average 3.34%) in leaves and from 2.52% to 7.56% (average 4.15%) in stems (Fig. 2f). For PheRDR6, editing efficiency ranging from 1.82% to 8.88% (average 4.89%) in leaves and from 1.28% to 7.78% (average 4.28%) in stems (Fig. 2g). Examination of the DlmRDR6 and PheRDR6 sequence near the sgRNA target site revealed various insertions and deletions (Indels), further confirming successful gene editing (Fig. 2h,i). Simultaneously, after 30 dpi with pBaMV-Cas9, we also performed PCR-RE detection on the cDNA of DlmRDR6 and PheRDR6 in the stems of D. latiflorus and P. edulis, respectively (Fig. S2e,f), which showed consistent mutations in DlmRDR6 and PheRDR6 genome DNA as well. To enhance BAMV-mediated gene editing utilizing the CRISPR/Cas system, we also explored co-expressed editing systems that expressed isolated Cas9 protein (complete version or split-Cas9) and sgRNA separately (Fig. S2g). We found that in the co-expression editing system, NbPDS gene editing could only be detected in the injected leaves of N. benthamiana, and that the split-Cas9 approach did not yield detectable editing (Fig. S2h). Recent studies have shown that variants of AsCas12f1, namely Ascas12f1-YHAM and Ascas12f1-HKRA, exhibit higher editing efficiency in rice (Ye et al., 2024). Thus, we constructed pBaMV-HKRA and pBaMV-YHAM vectors to target editing NbPDS. PCR-RE detection of NbPDS from symptomatic systemic leaves at 30 dpi indicated that AsCas12f1-HKRA had higher editing efficiency than AsCas12f1-YHAM (Fig. 2j). Amplicon deep sequencing showed that the pBaMV-HKRA vector-mediated editing efficiency ranged from 17.42% to 31.14% (average 26.27%), while pBaMV-YHAM editing efficiency ranged from 1.19% to 1.96% (average 1.72%). Although both variants of AsCas12f1 have significantly lower editing efficiency than Cas9 (Fig. 2k), the pBaMV-HKRA vector was able to induce large fragment deletions of target genes (Figs 2l, S2i; Notes S1), as confirmed by Sanger sequencing, suggesting significant potential for targeted DNA deletion using AsCas12f1-HKRA. In summary, utilizing this system, we successfully targeted and achieved DNA-free genome editing in D. latiflorus and P. edulis, demonstrating the potential of BaMV-mediated CRISPR/Cas9 for application in bamboo research. Although the current BaMV-mediated CRISPR/Cas9 system shows relatively low efficiency in genome editing in P. edulis and D. latiflorus, our study broadens the library of plant gene-editing tools based on positive-stranded RNA viruses and holds the potential to advance VIGE technology in plants. We aim to improve this system in bamboo by focusing on vector optimization and exploring other Cas proteins in future studies. Furthermore, due to the fact that BaMV virus can spread widely in the stem of D. latiflorus and P. edulis, BaMV-induced gene editing may facilitate a more rapid acquisition of gene-editing resources through stem regeneration. Notably, BaMV has a broad host range, which encompasses both monopodial bamboo and sympodial bamboo, significantly expanding the prospects for bamboo breeding and genetic improvement. This research was funded by the National Key R&D Program of China (2021YFD2200505), the National Natural Science Foundation of China (3237141305), the Natural Science Foundation of Fujian Province (2021J02027), the S&T Innovation (KFB23180) and the Forestry Peak Discipline Construction Project of Fujian Agriculture and Forestry University (72202200205). None declared. LG conceived and designed the research. LW, JY, YG, QW, HG and LZ performed experiments. ZZ and HZ performed the bioinformatics analyses. LW and LG prepared the manuscript. LW, JY and YG contributed equally to this work. The data that supports the findings of this study are available in the Supporting Information for this article. Supplementary figures related to the study can be found in Supporting Information Figs S1 and S2. For sequence information required for experimental integrity, refer to Table S1. Sanger sequencing results from targeted gene editing are available in Notes S1. Fig. S1 The gene structures and the editing efficiency, including data showing mutation types and frequencies. Fig. S2 The analysis of gene-editing efficiency in D. latiflorus and P. edulis. Notes S1 Results of Sanger sequencing. Table S1 Oligos, primers and relevant sequences used in this study. Please note: Wiley is not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. The New Phytologist Foundation remains neutral with regard to jurisdictional claims in maps and in any institutional affiliations.

Topics & Concepts