DNA‐free base editing in lettuce via in vitro transcribed base editors

A newly developed RNA-based adenine and cytosine base editing system achieves targeted and efficient A-to-G and C-to-T conversions in lettuce. This DNA-free base editing method has potential uses in crop breeding and biotechnology. Recent advances with DNA-free clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated nuclease 9 (Cas9) methods using ribonucleoproteins (RNPs) in protoplasts (Woo et al., 2015) or RNAs in callus (Zhang et al., 2016) can effectively modify the plant genome with minimal off-target effects and non-integrated DNA fragments, making them attractive for crop development (Chen et al., 2019). Base editors are composed of Cas9 nickase and adenine or cytosine deaminases to generate precise base conversions (A-to-G or C-to-T) at the desired target site (Anzalone et al., 2020). Base editing has been used to modify the plant genome in wheat and improve herbicide resistance using a transgene-free method (Zhang et al., 2019). Here, we set out to further improve the precision and efficiency of this strategy by introducing more recently engineered base editors using an RNA-based protoplast transfection method. To test the efficiency of base editing in our DNA-free system, we developed T7-promoter driven plasmids with either an engineered adenine base editor (ABE; TadA8e-nCas9(D10A)) (Richter et al., 2020) or cytosine base editor (CBE; hA3A-nCas9(D10A)-UGI) (Zong et al., 2018) (Figure S1). We also synthesized a number of single guide RNAs (sgRNAs) for our target genes of interest using a similar in vitro method. We delivered the base editor RNAs (ABE or CBE messenger RNA (mRNA) with sgRNA) into protoplasts from lettuce by polyethylene glycol (PEG)-mediated transfection before analyzing editing efficiency in protoplasts, calli, shoots and plants by targeted deep sequencing (Figure 1A). An efficient RNA-based adenine (ABE) and cytosine base editing (CBE) system in lettuce (A) Schematic overview of the generation of DNA-free base edited plants. (B) Target sequence in LsFT for DNA-free adenine base editing. The Y85H amino acid substitution confers a late-flowering phenotype. (C) Frequencies of A-to-G conversions at LsFT in lettuce protoplasts transfected with the ABE messenger RNA (mRNA) and single guide RNA (ABE-sgRNA) compared to mock-treated. (D) Overall A-to-G base editing frequencies (mean ± SEM) in ABE-sgRNA lettuce calli (n = 31) and shoots (n = 34) measured by targeted deep sequencing. (E) (left panel) Image of lettuce plants after growth in long-day conditions reveal delayed flowering in ABE-sgRNA-base edited plants compared to mock-treated lettuce. Scale bar = 5 cm. (right panel) Sanger sequencing and targeted deep sequencing show A-to-G conversion at LsFT indicative of the Y85H amino acid change. (F) Target sequence in LsALS for DNA-free cytosine base editing. Amino acid substitution (P184F) in LsALS confers herbicide resistance. (G) Frequencies of C-to-T conversions at LsALS in lettuce protoplasts transfected with the CBE-sgRNA compared to mock-treated. (H) Images of calli from herbicide resistance test using regeneration medium containing chlorsulfuron. DAI, days after incubation. (I) Overall C-to-T base editing frequencies (mean ± SEM) in CBE-sgRNA-treated plants grown with (+CS) or without chlorsulfuron (-CS). -CS calli (n = 19), -CS shoots (n = 12), +CS calli (n = 8) and +CS shoots (n = 7). (J) (left panel) Image of herbicide resistance test in CBE-sgRNA-base edited T1 generation compared to mock-treated T1 lettuce. (right panel) Sanger sequencing show C-to-T conversion at LsALS indicative of P184F amino acid change of T1 generation lettuce. (C), (G) Substitution efficiencies (mean ± SEM) were calculated from five independent experiments (n = 5). ****P < 0.0001 by two-tailed Student's t-test. (E, J) The arrows indicate the substitute nucleotides in the chromatograms. The protospacer adjacent motif sequence and sgRNA are shown in light blue and pink, respectively. (H, J) Scale bar = 1 cm. We first examined the DNA-free adenine base editing system in the lettuce flowering locus T (LsFT) by inducing late flowering to prolong the period of vegetative growth through Y85H amino acid substitution (Kang et al., 2018). We co-transfected lettuce protoplasts with ABE mRNA and sgRNA (ABE-sgRNA) to target this mutation (Figure 1B). The ABE-sgRNA induced A-to-G conversions in the deamination window at the LsFT locus at efficiencies of up to 6.24% in lettuce protoplasts through targeted deep sequencing (Figures 1C, S2). To investigate whether these edits were maintained during regeneration, we measured base editing efficiency in lettuce calli and shoots regenerated from ABE-sgRNA transfected protoplasts (Figures 1D, S3). In calli, A-to-G conversions were observed in 0.65% to 37.6% of samples sequences, and 0.50% to 50.0% in shoots. Impressively, ABE-sgRNA induced high-efficiency A-to-G conversions in lettuce calli and shoots without antibiotic selection, and genomic edits were transmitted to the next generation (T1; Figure S4). When ABE-sgRNA shoots were grown under long-day conditions, these plants revealed a late-flowering phenotype compared to controls, providing further evidence that edits were preserved during regeneration (Figure 1E). Taken together, these results show that ABE-sgRNA can effectively modify the plant genome without antibiotic selection. Next, we tested a cytosine base editor to modify lettuce acetolactate synthase (LsALS), a key enzyme in the biosynthesis of branched-chain amino acids. Several single nucleotide polymorphisms (SNPs) in the ALS locus, have been shown to confer sulfonylurea (SU) herbicide resistance in plants (Powles and Yu, 2010). The conversion of Pro 184 in LsALS to Phe leads to SU herbicide-resistance, and is therefore a high value target for improving yield in agriculture (Figure 1F) (Zhang et al., 2019). We co-delivered the CBE mRNA and sgRNA (CBE-sgRNA) into lettuce protoplasts using PEG-mediated transfection. Targeted deep sequencing revealed that CBE-sgRNA induced C-to-T conversions at efficiencies of up to 12.6% within the deamination window in lettuce protoplasts (Figures 1G, S5). To investigate whether the LsALS mutation could confer resistance to herbicide, we transferred the micro-calli regenerated from CBE-sgRNA-treated protoplasts to regeneration medium containing chlorsulfuron herbicide (Figure 1H). Mock-treated calli turned transparent white (dead) by exposure to herbicide. In contrast, CBE-sgRNA-treated regenerable calli retained their green color and generated shoots, indicating resistance to chlorsulfuron by target cytosine conversions with amino acid change. We analyzed CBE editing efficiencies in calli and shoots and observed C-to-T conversions in 0.35%–20.8% of calli and 0.11%–14.2% of shoots without chlorsulfuron treatment (Figures 1I, S6). Editing efficiencies further increased in herbicide-resistant lettuce calli of up to 96.4% and shoots of up to 51.2% (Figure S7). We next examined the germination test on media containing chlorsulfuron (SU). Sanger sequencing and targeted deep sequencing showed that C-to-T substitution was maintained in T1 (Figures 1J, S8). These results indicated that CBE-sgRNA transfection led to highly efficient, DNA-free cytosine base editing in plants, particularly when combined with herbicide selection. We also tested base editing in two additional target region of lettuce phytoene desaturase (LsPDS) and LsALS genes. ABE-sgRNA targeted to the LsPDS gene catalyzed the conversion of target adenine with A-to-G conversion frequencies of up to 3.53% in lettuce protoplasts, as measured by targeted deep sequencing (Figure S9). Likewise, CBE-sgRNA induced C-to-T conversions at the LsALS locus (Ala 109) at efficiencies of up to 8.15% within the deamination window in lettuce protoplasts (Figure S10). These results indicated that ABE-sgRNA and CBE-sgRNA can induce A-to-G and C-to-T substitutions efficiently and with high fidelity at multiple loci in lettuce. In summary, we have established an effective RNA-based adenine and cytosine base editing system in plants. Our ABE-sgRNA induced base editing within the LsFT target window in lettuce calli and shoots at frequencies of up to 37.6% and 50.0%, respectively. We also observed the late-flowering phenotype associated with the target Y85H amino acid change, indicating the A-to-G conversion was successful. Our CBE-sgRNA strategy led to C-to-T conversions within the LsALS target window in lettuce calli and shoots at frequencies of up to 20.8% and 14.2%, respectively. Herbicide treatment further increased base editing efficiencies of up to 51.2%. Due to consecutive Cs within the target window, cytosine base editing occurred at positions 6–10 of the protospacer. In this study, the CBE showed a broad window range and high efficiency, but it was found to have lower precision than the ABE. To evaluate the base conversions and efficiency within the target region, targeted deep sequencing analysis using protoplast transfection is required. Protoplast-based CRISPR-Cas9 gene editing induces a low incidence of chimerism (Lin et al., 2018). The base editors used in this study are also presumed to induce some chimerism while maintaining efficiency in cell division and plant development. We recommend using next generation lines that have been validated through Sanger sequencing or next generation sequencing. Our DNA-free base editing system could also be applied to other genome editing tools such as prime editors (Anzalone et al., 2020). This system provides a safe, precise and efficient genome editing strategy to improve crop breeding and biotechnology. This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea Government (MSIT) (No. RS-2022-NR072144), the Rural Development Administration under the New Breeding Technologies Development Program (No. RS-2024-00322224). The authors have declared no conflict of interest. E. L., Y. K., and B.-C. K. designed the study. E. L., Y. K., M. K., D. L. performed the experiments. B.-C. K. wrote the manuscript and supervised the project. All authors read and approved the contents of this paper. Additional Supporting Information may be found online in the supporting information tab for this article: http://onlinelibrary.wiley.com/doi/10.1111/jipb.13822/suppinfo Figure S1. Representation of the adenine and cytosine base editor for in vitro transcription Figure S2. A-to-G conversion efficiencies at LsFT in lettuce protoplasts Figure S3. The efficiencies and patterns of A-to-G conversions by adenine base editor (ABE) RNAs in lettuce calli (A) and shoots (B) at LsFT target region Figure S4. A-to-G substitution efficiencies and Sanger sequencing chromatograms in three independent LsFT T1 lines Figure S5. C-to-T conversion efficiencies at LsALS in lettuce protoplasts Figure S6. The efficiencies and patterns of C-to-T conversions by cytosine base editor (CBE) RNAs in lettuce calli (A) and shoots (B) at LsALS target region without herbicide selection Figure S7. The efficiencies and patterns of C-to-T conversions by cytosine base editor (CBE) RNAs in lettuce calli (A) and shoots (B) at LsALS target region with chlorsulfuron treatment Figure S8. The efficiencies and patterns (A) and genotypic segregation ratio (B) of C-to-T conversions by cytosine base editor (CBE) RNAs in germinated T1 seeds at LsALS target region Figure S9. Map of the LsPDS locus (A) and analysis of base editing efficiency and patterns in lettuce protoplasts (B) Figure S10. Map of the LsALS locus (A) and analysis of base editing efficiency and patterns in lettuce protoplasts (B) Table S1. Primers used in this study 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.