Targeted mutagenesis in Arabidopsis and medicinal plants using transposon‐associated TnpB

The programmable nuclease TnpB is significantly smaller than Cas9, can edit genes in medicinal plants, including Artemisia annua, Salvia miltiorrhiza, Scutellaria baicalensis, Isatis indigotica, and Codonopsis pilosula, and has potential uses in molecular breeding to enhance crop yield and quality. Medicinal plants produce valuable compounds, but often at low concentrations. Genome editing could be used to increase the production of valuable secondary metabolites in medicinal plants. The clustered regularly interspaced short palindromic repeat (CRISPR)/CRISPR-associated protein 9 (Cas9) system has emerged as a simple, widely used method for gene editing in plants. However, the identification of new endonucleases could improve its efficiency by expanding the range of target sequences, decreasing off-target effects, and, for smaller proteins, facilitating the delivery of genome-editing tools. The RNA-guided endonucleases Cas9 and IscB (encoded by a transposon in the IS200/IS605 family) may share a common ancestor with TnpB (Kapitonov et al., 2015), which is encoded by a small transposon in the IS200/IS605 family. TnpB has recently been used as an efficient tool for genome engineering in Escherichia coli and animal cells (Altae-Tran et al., 2021; Karvelis et al., 2021). TnpB can employ multiple guides to concurrently modify multiple genes (Wang et al., 2024), analogous to the mechanisms observed in Cas9 and IscB. However, its applicability for editing plant and non-human animal DNA has yet to be explored. Here, we successfully used TnpB for genome editing in the model plant Arabidopsis (Arabidopsis thaliana) and several medicinal plants. TnpB is an RNA-directed nuclease that can be guided by reRNA (right end element RNA), a 20-bp region matching the target gene (Karvelis et al., 2021). We developed a construct in which the reRNA was driven by the Arabidopsis U6 polymerase III promoter and TnpB was expressed under the control of the Ubiquitin1 (UBQ1) promoter (Figure 1A). In addition, we incorporated eukaryotic nuclear localization signals at both termini of TnpB to ensure its localization to the nucleus. To visualize the localization of TnpB, we fused TnpB with yellow fluorescent protein (YFP) to generate a YFP-TnpB construct. We expressed YFP-TnpB in Nicotiana benthamiana leaves and observed a clear nuclear localization (Figure S1), which is consistent with the localization of Cas9 protein (Nekrasov et al., 2013). Editing the plant genome using TnpB (A) The right end element RNA (reRNA):TnpB construct used for protoplast transformation, Agrobacterium tumefaciens-mediated transient transformation, and stable transformation in planta. (B) Diagram of the CHLI2 target site. (C) Agarose gel electrophoresis of the polymerase chain reaction products from target sites within the CHLI2 gene expressed in Arabidopsis protoplasts. Genomic DNA was digested with EcoRV (lanes 1–3); undigested genomic DNA is shown in lane 4. (D) Alignment of reads showing TnpB edits in CHLI2. The wild-type sequence is shown at the top. Bases that have been altered in edited plants are marked in red. The TAM (transposon-associated motif) is located next to the target sequence (at the 5′ end). Targeted genome editing of CYP71AV1 (E), SmTⅡAS (F), SbGUS (G), IsYUC2 (H), and SPSS2 (I) in Artemisia annua, Salvia miltiorrhiza, Scutellaria baicalensis, Isatis indigotica, and Codonopsis pilosula protoplasts, respectively. (J) TnpB shows indel-inducing activity in four targets, each spanning 20 nucleotides, within Arabidopsis genes ABCG39, NAP10, and GL3, as well as A. annua TLR1 (mean ± SD, n = 3 independent experiments). (K) Distribution of indel profiles for the target genes. The blue line represents deletions, while the red line represents insertions. The sequence within the green region represents the target, and the sequence within the pink region indicates the TAM. (L). Alignment of reads showing TnpB edits in CHLI2 (M). The target sequence recognized by reRNA:TnpB or Cas9 is highlighted in gray. The TAM is underlined, and DNA changes are shown in red. Next, we constructed a reRNA:TnpB vector to assess the genome-editing activity of TnpB in Arabidopsis. We generated a guide RNA by selecting a 20-bp sequence following the transposon-associated motif (TAM) "TTGAT" within the DNA sequence of Arabidopsis MAGNESIUM CHELATASE SUBUNIT I2 (CHLI2, At5g45930) (Figure 1B). To facilitate the identification of successful edits, we designed the target sequence to include the EcoRV restriction enzyme site, allowing detection of edits through restriction enzyme analysis (Nekrasov et al., 2013). To determine the mutagenesis efficiency of reRNA:TnpB in Arabidopsis protoplasts, we extracted genomic DNA and digested it with EcoRV. We amplified the digestion products by polymerase chain reaction (PCR) and sequenced the target region using Sanger sequencing (Table S1). By employing this technique, we effectively eliminated unedited target sites and selectively enriched the sites that underwent editing for subsequent target sequencing. Using wild-type plants as a control, we successfully amplified the edited target fragment from protoplasts expressing reRNA:TnpB, whereas neither TnpB nor the control vector was amplified (Figure 1C). These results indicate that TnpB was capable of editing the target sequence in the presence of the reRNA. Analysis of the CHLI2 PCR products revealed multiple mutations induced by TnpB, not only within the target sequence but also in adjacent regions (Figure S2). Deletions in CHLI2 were also detected. To verify the editing efficiency of our system for other genes, we also targeted Arabidopsis NAP10, ABCG39, and GL3 for editing. As expected, we observed both indels (insertions and deletions) and base pair substitutions in these target genes (Figure S3). To evaluate the editing rates in vivo, we evaluated Agrobacterium tumefaciens-mediated transient expression of the reRNA:TnpB constructs in Arabidopsis leaves. Mutations predominantly occurred in close proximity to the target sequence, consistent with our findings in protoplasts (Figure S4). These results provide evidence that TnpB is capable of editing CHLI2 in vivo. We investigated whether stable transformation of reRNA:TnpB could result in a similar level of gene editing in Arabidopsis. We generated transgenic Arabidopsis plants via A. tumefaciens-mediated transformation using the floral dip method. Approximately 50 T1 transgenic seedlings were obtained. Sequencing of the amplified target genes revealed many base substitutions near the target sequence (Figures 1D, S5), which is consistent with the findings of previous studies of TnpB-mediated genome editing in E. coli and human cells (Altae-Tran et al., 2021; Karvelis et al., 2021). The mutation efficiency was 14% (seven mutant plants out of 50 transgenic plants). However, all the edited plants showed a chimeric phenotype, suggesting that TnpB, similar to CRISPR/Cas9, induces editing in individual cells. Additionally, some short deletions and insertions were observed flanking the target sequence (Figure S3A, C). In the subsequent generation (T2), the chimerism mutants displayed a pale-green phenotype resembling that of other CHLI2 mutant lines (Figure S6) (Mao et al., 2013). The relatively short length of the target sequence (20 bp) enabled non-specific binding, resulting in off-target effects similar to those observed using CRISPR/Cas9. To assess the occurrence of off-target mutations in Arabidopsis, we searched for the 20-bp target sequences in the Arabidopsis TAIR10 genome (https://www.arabidopsis.org/). This search yielded 13 potential off-target sequences (Table S2), which shared 15 to 17 bp of identical sequences with the target CHLI2 sequence. Within these sequences, we identified no off-target mutations. To obtain homozygous mutants in the T1 generation, we used an egg cell-specific promoter (egg-cell pro) to drive Cas9 and TnpB expression. The editing efficiency of egg cell pro:Cas9 was approximately 17%; however, no homozygous mutant was obtained in the T1 generation of egg cell pro:Cas9 transgenic plants. The editing efficiency and ability to generate homozygous mutants is affected by many factors (Wang et al., 2015). We also failed to detect homozygous mutants in the egg cell pro:TnpB transgenic plant population, with a mutation efficiency of 13% (Figure 1L, M). We evaluated whether TnpB could be used to edit genes in the medicinal plant species Artemisia annua, Salvia miltiorrhiza, Scutellaria baicalensis, Isatis indigotica, and Codonopsis pilosula. We targeted CYP71AV1, SmTⅡAS, SbGUS, IsYUC2, and SPSS2 for editing in A. annua, S. miltiorrhiza, S. baicalensis, I. indigotica, and C. pilosula, respectively, and identified mutations in transiently transformed protoplasts by Sanger sequencing. Consistent with Arabidopsis, TnpB successfully edited the target genes in these species (Figure 1E–I). We identified insertions in S. miltiorrhiza (Figure 1F), but not in the other species. To assess the overall effects of TnpB-mediated gene editing, we performed high-throughput sequencing in Arabidopsis and A. annua. We transiently expressed the reRNA:TnpB vector, which we used to edit NAP10, ABCG39, and GL3 in Arabidopsis protoplasts and TLR1 in A. annua protoplasts. We extracted genomic DNA from the protoplasts and subjected it to high-throughput sequencing to detect indels in the targeted cleavage sites (Liu et al., 2019). In Arabidopsis, the efficiency of indel generation in NAP10, ABCG39, and GL3 edited by TnpB was 30%, 2%, and 15%, respectively (Figure 1J). Intriguingly, the efficiency of indel induction in NAP10 was higher than that previously reported in human cells (Karvelis et al., 2021) (Figure 1K). In A. annua, the efficiency of indel induction in TLR1 was approximately 2% (Figure 1J). Moreover, the highly edited sites were located outside the target regions, with lower deletion efficiencies within all four genes (Figures 1K, S7). These results suggest that the editing efficiency might be associated with gene structure. To further analyze the mutational profiles of the gene-edited plants, we examined 50-bp sequences that had acquired indels near the target sites, revealing a prevalence of deletions compared with insertions at the cleavage site (Figures 1K, S7), which is consistent with prior findings (Karvelis et al., 2021). Fewer deletions were detected near to the target sites than at more distant locations (Figures 1K, S7). We also observed base pair substitutions at the target sites. As shown in Figure S9A, the rate of C-to-G substitutions in TLR reached 33.07% in test 1. The efficiency of base pair substitutions was notably lower in ABCG39, NAP10, and GL3, falling below 3% (Figures S8, S9). Recent studies have indicated that TnpB induces significant indels in rice (Karmakar et al., 2024; Li et al., 2024), whereas base pair substitutions observed in Arabidopsis may be associated with relatively lower efficiency. Overall, TnpB guided by reRNA represents a suitable tool for genome editing in plants. Our findings demonstrate that the transposon protein TnpB is capable of inducing mutations in target genes in plants. TnpB is a relatively small gene (only 1,227 bp), which makes it highly amenable to manipulation and utilization for gene editing. Besides base changes, TnpB also can delete genomic fragments. The TAM site used by TnpB (TTGAT) is two bases longer than the protospacer adjacent motif (PAM) sites used by Cas9, but this site contains three more specific bases than the PAM of SpCas9 (NGG). In addition, since it is easy to create base pair substitutions using TnpB, this nuclease could be developed as a tool for generating base pair substitutions. Additional optimization is needed to perfect TnpB as a tool. Nevertheless, this small nuclease has the potential for use in molecular breeding to improve crop yield and quality. This work was funded by the National Key Research and Development Program of China (grant no. 2022YFC3501700), National Natural Science Foundation of China (32070332), and Shanghai Local Science and Technology Development Fund Program guided by the Central Government (YDZX20203100002948). We thank Tgene Biotech (Shanghai) Co., Ltd. for providing us with the high-throughput data. The authors declare no conflict of interest. Z.L. and W.C. conceived and designed the entire research plan; W.C., X.W., B.D, and S.F. performed most of the work; J.X. provided technical assistance; Z.L., W.C., and X.W. wrote the manuscript; and W.C. and L.Z. helped with the organization and editing. 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.13758/suppinfo Figure S1. Localization of TnpB is in plant nuclei and cytoplasm Figure S2. TnpB edits DNA in the protoplast of Arabidopsis Figure S3. TnpB creates editing in NAP10, ABCG39, and GL3 of Arabidopsis Figure S4. TnpB edits DNA in Arabidopsis leaves Figure S5. TnpB edits DNA in transgenic Arabidopsis plants Figure S6. The phenotype of right end element RNA (reRNA):TnpB transgenic plants Figure S7. Distribution of indel profiles in target genes is depicted. The blue line signifies deletions, while the red line denotes insertions. The sequence within the green region signifies the target, with the pink region indicating transposon-associated motif (TAM) Figure S8. The base pair substitution rate of TnpB in targets of 20 nt in length is illustrated in (A) and (B) Figure S9. The base pair substitution rate of TnpB in targets of 20 nt in length is illustrated in (A)and (B) Supplementary Table 1 Primer list. TTGAT represents transposon-associated motif (TAM) site Supplementary Table 2 Potential off-target sites 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.