An Agrobacterium-mediated base editing approach generates transgene-free edited banana Senne Van den Broeck1,2 , Yvan Ngapout1,2 , Bart Panis1,2,3 and Herv�e Vanderschuren1,2,4 1Laboratory of Tropical Crop Improvement, Department of Biosystems, KU Leuven, Willem de Croylaan 42, 3001, Heverlee, Belgium; 2KU Leuven Plant Institute (LPI), Kasteelpark Arenberg 31, 3001 Heverlee, Leuven, Belgium; 3Alliance of Bioversity International and CIAT, KU Leuven, Willem de Croylaan 42, 3001, Heverlee, Belgium; 4Plant Genetics Laboratory, TERRA Teaching and Research Center, Gembloux Agro-Bio Tech, University of Li�ege, Passage des D�eport�es 2, 5030, Gembloux, Belgium Author for correspondence: Herv�e Vanderschuren Email: herve.vanderschuren@kuleuven.be Senne Van den Broeck Email: senne.vandenbroeck@kuleuven.be Received: 16 December 2024 Accepted: 14 February 2025 New Phytologist (2025) doi: 10.1111/nph.70044 Key words: acetolactate synthase, banana (Musa spp.), co-editing, CRISPR/Cas9, gene editing, transgene-free. Summary � Genome editing for the development of improved varieties is supported by the possibility of segregating out the editor T-DNA cassette after genome editing in many crop species. Removal of the T-DNA cassette prevents potential continuous editing activity in the trans- formed plant and furthermore facilitates regulatory approval. While transgene outcrossing of exogenous sequences is possible for many crops, this is not the case for vegetatively propa- gated and sterile crops, such as Cavendish bananas. Therefore, gene editing techniques lead- ing to transgene-free edited plants are essential to untap the potential of genome editing for those crops. Here, we present a method for transgene-free gene editing in sterile banana (Musa spp.) through a co-editing strategy. � A novel Agrobacterium tumefaciens-mediated transgene-free gene editing approach com- bining embryogenesis and chlorsulfuron selection was established in sterile banana and vali- dated through whole genome sequencing. � Editing of the acetolactate synthase (MaALS) genes in banana using a plant base editor allows effective selection of edited plants. Moreover, transgene-free plantlets were regener- ated with mutations at two target sites, indicating that the strategy can be used to target mul- tiple genomic sites. � The presented method allows for efficient transgene-free gene editing and represents the first report of a co-editing strategy in sterile crop species. Introduction In the last decade, genome editing has revolutionized crop breed- ing by significantly reducing the time required for the introduc- tion of improved traits in elite germplasms (Razzaq et al., 2021). The clustered regularly interspaced short palindromic repeat (CRISPR)-/CRISPR-associated nuclease 9 (Cas9) system facili- tates the rapid and cost-effective targeting of plant genome sequences to introduce mutations, with the possibility to increase their resilience to pests and diseases, abiotic stress factors or improving nutritional traits (Li et al., 2024). The initial develop- ment of CRISPR/Cas9 in plants utilized Agrobacterium tumefa- ciens for the genomic integration of a transfer DNA (T-DNA) expressing the guide RNA (gRNA) and accompanying protein (Li et al., 2013; Nekrasov et al., 2013; Shan et al., 2013). How- ever, the Agrobacterium-mediated introduction of transgenes into plant genomes is random and can result in unpredictable and unstable outcomes (Jaganathan et al., 2018). Moreover, the use of genetically modified organisms remains controversial in public opinion and is subject to costly approval procedures in many parts of the world (Kennedy & Thigpen, 2020; Levi, 2022; Vanderschuren et al., 2023). Hence, transgenesis still limits the commercialization of edited crops (Turnbull et al., 2021; Ahmad et al., 2023). However, removal of the transgene can be achieved for most crops through transgene outcrossing (Gao, 2021) and facilitates the generation of transgene-free edited plants in the T1 generation. Removal of the T-DNA cassette, moreover, prevents potential continuous editing activity in the transformed plant as well. Because outcrossing of transgenes remains challenging for vegetatively propagated and sterile crops, methodological devel- opment has to be focussed on approaches circumventing foreign DNA integration to generate transgene-free gene-edited plants immediately in the T0 generation (X. Huang et al., 2023). For example, DNA-free gene editing methods using mRNA transcripts (Zhang et al., 2016; Liang et al., 2018) or ribonucleo- protein complexes, consisting of preformed Cas9 and single-guide RNAs (sgRNAs) (Svitashev et al., 2016; Liang et al., 2017, 2018; Zong et al., 2018; Li et al., 2022), have been established to edit plant genomes. These can be delivered through a variety of DNA-free gene editing techniques, such as particle bombardment, often of immature embryos, or protoplast electroporation or polyethylene glycol transformation (Woo et al., 2015; H. Kim et al., 2017; Andersson et al., 2018; Banakar et al., 2022). Another approach consists of using transient � 2025 The Author(s). New Phytologist� 2025 New Phytologist Foundation. New Phytologist (2025) 1 www.newphytologist.com Research https://orcid.org/0000-0002-4493-5305 https://orcid.org/0000-0002-4493-5305 https://orcid.org/0000-0002-1719-138X https://orcid.org/0000-0002-1719-138X https://orcid.org/0000-0001-6717-947X https://orcid.org/0000-0001-6717-947X https://orcid.org/0000-0003-2102-9737 https://orcid.org/0000-0003-2102-9737 mailto:herve.vanderschuren@kuleuven.be mailto:senne.vandenbroeck@kuleuven.be http://crossmark.crossref.org/dialog/?doi=10.1111%2Fnph.70044&domain=pdf&date_stamp=2025-04-01 exogenous DNA expression. Exogenous DNA is delivered through particle bombardment (Zhang et al., 2016; Hamada et al., 2018; Zong et al., 2018) or protoplast transformation (Andersson et al., 2017; Lin et al., 2018; Hsu et al., 2024), and the subsequent transient expression allows editing of the genome. In addition to the above-mentioned delivery methods, transient expression can also occur from T-DNA delivered through A. tumefaciens. After T-DNA delivery into the host cell as a single-stranded molecule (Albright et al., 1987), double-stranded DNA is generated independently of T-DNA integration, enabling transcription of encoding genes (Zhuobin Liang & Tzfira, 2013). It has been shown that mutations can also be intro- duced by transiently expressed T-DNA (Chen et al., 2018; Zong et al., 2018; R. Zhang et al., 2019; B�anfalvi et al., 2020; X. Huang et al., 2022, 2023; Jia et al., 2024), without integration in the host genome. However, as no transgenic DNA is integrated in the above-mentioned methods, only the selection of edited cells is extremely difficult, laborious and expensive (Kocsisova & Con- eva, 2023). While plants derived from protoplasts are mostly the result of a single-cell event (Reed & Bargmann, 2021), this is not the case for plants derived from other, multicellular, tissues, often resulting in chimerism. Methods allowing the selection of edited cells and tissues are required to avoid the regeneration of chimeric plants that are partially edited. Moreover, contrary to DNA-free genome-editing techniques, Agrobacterium-based tran- sient expression of CRISPR/Cas9 not only involves the selection of those plants harbouring mutations but also has the extra criter- ion that selected regenerants need to be transgene-free. One of the promising selection markers to select transformed and mutated cells and tissues is acetolactate synthase (ALS ) or acetohy- droxyacid synthase (AHAS ), an essential gene in the synthesis of branched amino acids (Yu & Powles, 2014). Acetolactate synthase has been used extensively as a proof-of-concept target site for gen- ome editing, given its conservation in the plant kingdom, as well as an easily observable phenotype upon mutation (Hussain et al., 2021). Indeed, ALS can result in sulfonylurea herbicide resistance by mutations at various distinct target sites in its coding sequence, with the most frequently observed and used mutation being Pro197 (Arabidopsis thaliana residues), which can be modi- fied into at least 11 other amino acids to result in chlorsulfuron resistance (Heap, 2024). It is therefore a suitable target to be used as an immediate selection marker to confirm the presence of edits in proof-of-concept studies, as previously demonstrated in tomato, tobacco, potato, citrus and maize (Svitashev et al., 2016; Danilo et al., 2019; Veillet et al., 2019; Alqu�ezar et al., 2022; X. Huang et al., 2022, 2023; Jia et al., 2024). Evidently, the most desired mutations are not easily selectable, requiring vast screen- ing methods to identify plants containing the correct and desired mutation(s). However, it has recently been shown that using edits in ALS, other secondary mutations at desired target sites can be enriched, thereby allowing efficient co-editing of your required alleles (X. Huang et al., 2023; Jia et al., 2024). Nevertheless, regeneration through organogenesis from explants such as leaf often results in chimeric plants. In contrast to shoot organogen- esis, embryogenesis from embryogenic cell suspensions (ECS) has a lower chance of resulting in chimerism as individual clumps are mostly seen as a single entity (Bertsch et al., 2005; Bhatia & Bera, 2015). Generating transgene-free gene-edited plants is especially important in vegetatively propagated and sterile crops, such as banana. Banana is the most important fruit crop world-wide, with an annual global production of c. 135 million tonnes (FAO, 2023). A single variety, Cavendish, represents over 40% of global banana production, and due to the widespread mono- cropping practices, it suffers from extreme levels of genetic vul- nerability (Drenth & Kema, 2021). As a result, several major diseases have emerged and have spread globally, with Fusarium oxysporum f. sp. cubense tropical race 4 now also reaching the sta- tus of pandemic (Zorrilla-Fontanesi et al., 2020; Turrell, 2024). Other major challenges include banana bunchy top virus, black sigatoka, nematodes and Xanthomonas wilt (Drenth & Kema, 2021; Justine et al., 2022; Tripathi et al., 2024). Despite some successes in generating improved cultivars through classical breeding for smallholder farmers (Ortiz, 2013; Tushemereirwe et al., 2015), the sheer scale and duration (up to 10–20 yr) of such a project make it an undesirable pathway for cultivar innovation. New traits are mostly introduced through A. tumefaciens-mediated transformation (P�erez Hern�andez et al., 2006; Remy et al., 2013; Dale et al., 2017; Tripathi et al., 2019; Zorrilla-Fontanesi et al., 2020), although transgenic bananas have also been regenerated through particle bombard- ment (S�agi et al., 1995; Matsumoto et al., 2002; Dong et al., 2020). However, these techniques have in common the integration of exogenous DNA as the main bottleneck. While strategies have been established that allow for recombination-induced deletion of either the selectable marker (Kleidon et al., 2020) or all transgenes (Hu et al., 2023) inte- grated into the banana genome, they inevitably leave scars at the integration site, which could represent a major burden for their regulation in different parts of the world, including the Eur- opean Union. The possibility to utilize protoplasts for CRISPR/Cas9-mediated genome editing of desired target sites has been investigated (Wu et al., 2020; Zhao et al., 2022), but regeneration of banana plants from protoplasts remains challen- ging, and there is so far no report of successful production of edited banana plants using protoplast systems. Here, we present a simple and efficient method for the genera- tion of transgene-free gene-edited banana plants (Musa acumi- nata cultivar (cv) Williams). We show that editing both of the ALS genes in banana is a viable strategy for the selection of edited plants. Moreover, edits in two distinct genes in a transgene-free way can be achieved following this strategy. Materials and Methods Plant material ECS of the Musa acuminata cv ‘Williams’ (AAA) (subgroup Cavendish) were initiated from ITC0570 (International Musa Transit Centre) and maintained in liquid ZZ medium as described previously (Strosse et al., 2003). Suspensions were New Phytologist (2025) www.newphytologist.com � 2025 The Author(s). New Phytologist� 2025 New Phytologist Foundation. Research New Phytologist2 14698137, 0, D ow nloaded from https://nph.onlinelibrary.w iley.com /doi/10.1111/nph.70044 by C ochraneItalia, W iley O nline L ibrary on [15/04/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense cultured on a rotary shaker (70 rpm) at 26 � 2°C under contin- uous light of 50 lE m�2 s�1 and subcultured at 2-wk intervals. Acetolactate synthase sequence analysis Acetolactate synthase genes in banana were identified through blast with the Arabidopsis thaliana (L.) Heynh. ortholog CSR1 (AT3G48560), also known as AHAS or ALS. Genes identified in the reference genome Musa acuminata DH Pahang (v.4.3) (Belser et al. (2021), Banana Genome Hub) were used as a starting prompt for BLASTN (Camacho et al., 2009) for further phylogenetic studies in otherMusaceae species with available fully sequenced genomes (Supporting Information Table S1) (Zorrilla-Fontanesi et al., 2016; Rouard et al., 2018; Wang et al., 2019; Belser et al., 2021; Wang et al., 2022; H.-R. Huang et al., 2023; Dussert et al., unpublished; Ziwei et al., unpub- lished). Haplotypes for Williams were resolved by cloning (CloneJet PCR Cloning Kit, Cat. no. K1231; Thermofisher Scientific, Waltham, MA, USA) and colony sequencing. Phylo- geny mapping was performed through MUSCLE (Edgar, 2004) and PHYLIP Neighbour Joining (Felsenstein, 1989). Arabidopsis transformation with bananaMaALS genes To validate the functionality of the banana MaALS genes, Macma4_06_g18410 and Macma4_10_g15750 were codon- optimized for A. thaliana (Twist Bioscience, South San Fransisco, CA, USA). The proline 186 or 181 residue was adapted to either a leucine or phenylalanine. Sequences were hybridized into pGGC000 (Lampropoulos et al., 2013) between KpnI and BamHI restriction sites through Gibson assembly using the GenBuilderTM Cloning Kit (Genscript, Piscataway, NJ, USA) to form a functional C-cassette for Greengate assembly. Codon- optimized genes were placed under a CaMV35S promoter and terminator through Greengate assembly together with plasmids pGGA004, pGGB003, pGGD002, pGGE001, pGGF007 and pGGZ003 according to Lampropoulos et al. (2013). Plasmids were electroporated into A. tumefaciens GV3101, and A. thaliana Col-0 was transformed through floral dip. Plasmid generation For banana, a cloning vector was designed containing a cytosine base editor (addgene #98160; Zong et al., 2017), with the possi- bility of integrating two sgRNAs. The base editor was amplified with primers PBE_pETKUL21_F and PBE_pETKUL21_R (Table S2) to integrate SpeI and BamHI restriction sites, after which the fragment was integrated into pETKUL21 (pCam- bia1380+maize Ubiquitin promoter, unpublished results) through restriction-ligation at mentioned sites, generating plas- mid pETKUL21_PBE. On pYPQ131C (addgene #69284; Low- der et al., 2015), BbsI overhangs were introduced bordering the sgRNA module to generate two independent sgRNA entry mod- ules, with primers pENTR1_F1 and pENTR1_R1, and pENTR_F2 and pENTR_R2, for pENTR1 and pENTR2, respectively. Fragments were incorporated into TOPO plasmids using Zero BluntTM TOPOTM PCR Cloning Kit (Cat. no. 451245; Thermofisher Scientific). sgRNAs were designed using CRISPOR (Concordet & Haeussler, 2018). An extra G was incorporated at the 50 end of the sgRNAs to increase efficiency (Belhaj et al., 2013). sgRNAs were integrated into entry modules by annealing two comple- mentary oligos, each harbouring a 4-bp overhang mimicking Esp3I overhangs in pENTR1 and pENTR2. 100 ng of annealed oligos was mixed with 50 ng of pENTR1 or pENTR2, together with 10 U Esp3I (Cat. no. R0734S; New England Biolabs (NEB), Ipswich, MA, USA), 400 U T4 DNA ligase (Cat. no. M0202S; NEB), 109 CutSmart buffer and 1.5 ll 10 mM ATP in a 15 ll reaction. The reaction was incubated for 30 cycles at 37°C for 3 min and 16°C for 3 min, followed by 5 min at 50°C, 5 min at 80°C and an infinite hold at 12°C. Ligation products were heat-shock-transformed in E. coli, and positive clones were selected with PCR through the reverse sgRNA pri- mer, together with M13 reverse primer. For the integration of the sgRNAs in destination vector pET- KUL21_PBE, pETKUL21_PBE was cut with BamHI-HF (Cat. no. R3136S; NEB), HindIII-HF (Cat. no. R3104S; NEB) and the addition of Quick CIP (Cat. no. M0525S; NEB) and puri- fied using the GeneJET PCR purification kit (K0702; Thermo- fisher Scientific). pENTR1 and pENTR2 (containing the sgRNAs) were cut with BbsI-HF (Cat. no. R3539S; NEB) and purified using the GeneJET gel extraction kit (Cat. no. K0692; Thermofisher Scientific). Components were ligated with T4 DNA ligase, and the ligation products were heat-shock- transformed into Escherichia coli. Positive colonies were selected by PCR with primers pDEST_F and the reverse sgRNA primer. In vitro cleavage assay To validate the functionality of the designed sgRNAs, an in vitro cleavage assay was performed. Designed sgRNAs were cloned under a T7 promoter in vector pT7-gRNA (addgene #46759; Jao et al., 2013) for in vitro transcription. In vitro transcription was performed as in Zhen Liang et al. (2018), with an initial 3-h incubation step at 37°C for RNA production. sgRNAs were mixed with purified Cas9 protein (Cat. no. M0386T; NEB) according to the manufacturer’s protocol and incubated for 10 min at 25°C, before adding substrate DNA. Substrate DNA was amplified with primers bALS6_F1 and bALS6_R4 for MaALS6 and bALS10_F3 and bALS10_R4 for MaALS10, respectively. After cleavage for 15 min at 37°C, 1 ll Proteinase K (Cat. no. 8107S; NEB) was added, and the samples were incu- bated at room temperature for 10 min, after which the samples were loaded on gel. Banana transformation and selection on chlorsulfuron Agrobacterium tumefaciens EHA105 was used for the transforma- tion of banana cultivar Williams. Before transformation, A. tume- faciens was plated on YM medium (0.1 g l�1 NaCl, 0.2 g l�1 MgSO4�7H2O, 0.5 g l�1 K2HPO4�3H2O, 0.4 g l�1 yeast extract, 10 g l�1 mannitol, 13 g l�1 bacto agar, pH 7.0) at � 2025 The Author(s). New Phytologist� 2025 New Phytologist Foundation. New Phytologist (2025) www.newphytologist.com New Phytologist Research 3 14698137, 0, D ow nloaded from https://nph.onlinelibrary.w iley.com /doi/10.1111/nph.70044 by C ochraneItalia, W iley O nline L ibrary on [15/04/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense 28°C for 2 d, after which single colonies were inoculated in YEP medium (10 g l�1 yeast extract, 10 g l�1 peptone, 5 g l�1 NaCl) and incubated at 28°C and 210 rpm for 30 h. Banana transformation was performed as published previously (P�erez Hern�andez et al., 2006; Kov�acs et al., 2013). After 6 d on nonse- lective ZZ semisolid medium, cocultivated samples were trans- ferred to ZZ containing 200 lg ml�1 timentin supplemented with either 50 lg ml�1 hygromycin or 40 lg l�1 chlorsulfuron. Clumps were grown for 8–14 wk with refreshment of the med- ium every 2 wk. After 8 to 14 wk, clumps growing on both hygromycin and chlorsulfuron were transferred to RD1 regenera- tion medium (Strosse et al., 2003) containing 40 lg l�1 chlor- sulfuron and 200 lg ml�1 timentin and subcultured every month for 1–3 months, after which they were transferred to RD2 (Strosse et al., 2003) without antibiotics. After subculturing for 1–3 months, lines were transferred to Reg medium (P�erez Hern�andez et al., 2006) and finally placed in the light at 100 lmol m�2 s�1. DNA purification DNA was purified through cetyltrimethylammonium bromide (CTAB) extraction. A small piece of cell clump or leaf (for whole genome sequencing (WGS) analysis) was crushed and resus- pended in 400 ll of cetyltrimethylammonium bromide buffer (10 mM Tris–HCl pH 7.5, 0.7 M NaCl, 10 mM EDTA, 1 g l�1 CTAB). Samples were incubated at 60°C for 1 h with intermittent inversion, after which 200 ll of ice-cold chloroform was added and mixed for 1 min after the samples had returned to room temperature. After centrifugation at 1500 g for 10 min, supernatant was transferred to a new tube and 440 ll of ice-cold isopropanol was added. Samples were incubated at �20°C for 20 min, followed by centrifugation at 1500 g for 15 min. Super- natant was discarded, and the pellet was resuspended in 70% ethanol, followed by another identical centrifugation step. Etha- nol was removed and evaporated, after which the genomic DNA was resuspended in MQ water. WGS and data analysis To confirm the absence of T-DNA in regenerated lines, purified DNA samples were first subjected to a PCR targeting the T- DNA. Samples were verified for the presence of the T-DNA with primers T-DNA-check_2F/R (PP1) and T-DNA-check_5F/R (PP2) (Table S3). Primer pairs Sanger_ALS6_F/R and Sanger_ALS10_F/R served as a control to warrant the quality of the purified DNA. Moreover, for 10 samples per transformed plasmid, one amplification with the latter amplicons was sent for Sanger sequencing to verify mutation ratios at on-target sites. Allelic variability was always interpreted conservatively, to avoid overestimating mutation rates. Where the same chromatogram could be the result of multiple mutation profiles on all alleles, allelic variability was interpreted to contain the least amount of edited alleles. Six randomly selected putative transgene-free lines were sent for WGS at 109 genome coverage per line to verify the absence of T-DNA (Suzuki et al., 2008; Willems et al., 2016). Addition- ally, a control plant from the cultivar Williams from the in vitro stock at the International Musa Transit Centre (KU Leuven, Bel- gium) and one regenerated from cell suspension but without transformation were sent along. The 150-bp paired-end reads WGS data were generated using the Illumina NovaSeq 6000 platform by GenomeScan (Leiden, the Netherlands). Image analysis, base calling and quality check were performed with the Illumina data analysis pipeline RTA v.3.4.4 (Illumina Inc, 2024) and BclConvert v.4.2.4 (Illumina Inc, 2023). The sequences of the eight samples were checked with FASTQC v.0.11.9 for final quality inspection. Subsequently, the high-quality paired-end short genomic reads (≥ Q30) were mapped with BWA-MEM v.0.7.17 (Li & Durbin, 2009) to the reference genome of Musa acuminata (DH Pahang v.4; https://banana-genome-hub.southgreen.fr/filebrowser/download/ 522). The alignment files were sorted with SAMTOOLS v.1.13 (Li et al., 2009), and duplicates were marked with PICARD v.2.18.23 (Broad Institute, 2009). The target site mutations were visually inspected using IGV v.2.17.4 (Robinson et al., 2011). To determine whether the edited lines contain fragments from the plasmids, the high-quality paired-end short genomic reads were mapped to the reference plasmid sequences (pETKUL21- PBE-ALS6 or pETKUL21-PBE-ALS6+10) using BWA-MEM v.0.7.17 (Li & Durbin, 2009) and visualized with IGV v.2.17.4 (Robinson et al., 2011) (Table S4). Using BLASTN (Camacho et al., 2009), putative integration sites for the WGS reads were verified by blasting the WGS reads mapping to the plasmid sequences pETKUL21-PBE-ALS6 or pETKUL21-PBE-ALS6+10 back to the Musa acuminata genome (DH Pahang v4). For the transgenic sample ALS6:21-3, integration in the genome was checked using TC-hunter (B€orjesson et al., 2022), which utilizes read alignment information to detect breakpoints in the plasmid and integration sites in the genome. Potential off-target sites were determined using CRISPOR and CasOFFinder (Bae et al., 2014; Concordet & Haeussler, 2018) and were investigated in WGS samples for off-target mutations (Table S5). The target sites were screened for C-to-D mutations within the editing window of the base editor (Bases 3–10) or frameshift mutations in the entire protospacer. Off- target mutations were considered valid if present in at least two independent reads. Results To utilise chlorsulfuron as a selective agent for gene editing (Veillet et al., 2019), it is essential to first investigate whether banana cv Williams is naturally resistant. ECS of Williams were used to investigate its natural resistance, as this is the tar- get tissue of choice for Agrobacterium-mediated transformation. The natural resistance to chlorsulfuron was first profiled using a range of chlorsulfuron concentrations (Figs 1a, S1). While cell weight did increase slightly for all chlorsulfuron concentra- tions after 2 wk, growth stagnated shortly after, even for con- centrations as low as 10 lg l�1. After 8 wk on selective medium, weight had increased between two- and threefold for New Phytologist (2025) www.newphytologist.com � 2025 The Author(s). New Phytologist� 2025 New Phytologist Foundation. Research New Phytologist4 14698137, 0, D ow nloaded from https://nph.onlinelibrary.w iley.com /doi/10.1111/nph.70044 by C ochraneItalia, W iley O nline L ibrary on [15/04/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://banana-genome-hub.southgreen.fr/filebrowser/download/522 https://banana-genome-hub.southgreen.fr/filebrowser/download/522 https://banana-genome-hub.southgreen.fr/filebrowser/download/522 https://banana-genome-hub.southgreen.fr/filebrowser/download/522 https://banana-genome-hub.southgreen.fr/filebrowser/download/522 https://banana-genome-hub.southgreen.fr/filebrowser/download/522 all chlorsulfuron concentrations, while weight increased more than 20-fold for cells plated on nonselective medium. A con- centration of 10 lg l�1 of chlorsulfuron performed similar to the positive control hygromycin at a concentration of 50 lg ml�1. Two banana MaALS genes were identified in the reference genome of Musa acuminata DH Pahang (v.4.3) (Banana Gen- ome Hub) (Droc et al., 2013; Belser et al., 2021), Mac- ma4_06_g18410 and Macma4_10_g15750, and they were abbreviated to MaALS6 and MaALS10, based on their respective chromosomes. This duplication event is conserved among the Musaceae family, and their genomic locations are conserved among the Musa genus (Fig. S2). MaALS6 and MaALS10 showed 71% and 77% protein identity, respectively, to their Ara- bidopsis thaliana ortholog CSR1 (AT3G48560), also known as AHAS or ALS. Transcriptomic analysis indicated that both genes are active, with MaALS6 consistently transcribed at a higher level in comparison withMaALS10 (Fig. S3). The genes MaALS6 and MaALS10 were amplified from ECS of cv Williams and mapped to the reference genome. Allelic differences at both gene and protein levels were identified (Genbank PQ304390–PQ304395) but were not present in the target region, that is the protospacer, confirming the absence of natural resistance in Williams ECS. Residues Pro186 and Pro181 were identified as homologous to Pro197 in A. thaliana, for MaALS6 and MaALS10, respectively. Analysis of the tar- geted region surrounding these proline residues showed that it is conserved among Musaceae (Fig. S2). MaALS6 and MaALS10 were codon-optimized for A. thaliana, and three versions of each gene were generated. MaALS6 or MaALS10 was either introduced into A. thaliana as is, or the above-mentioned pro- line residues were altered to either leucine or phenylalanine (Pro186Leu and Pro186Phe for MaALS6, Pro181Leu and Pro181Phe for MaALS10). In other species, these mutations were shown to be a gain-of-function, resulting in chlorsulfuron resistance while maintaining the enzyme function. Fig. 1 Construct overview for cytosine base editing in banana cultivar ‘Williams’ (Musa spp.). (a) Resistance of nonedited embryogenic cell suspensions of the cultivar Williams towards different concentrations of chlorsulfuron compared with 50 lg ml�1 hygromycin. Whiskers represent SE. (b) Single-guide RNAs (sgRNAs) designed to targetMacma4_06_g18410 (MaALS6) andMacma4_10_g15750 (MaALS10) are indicated in purple beneath the gene sequence. The targeted proline residue is indicated in yellow. Protospacer adjacent motif is indicated in red. Nucleotide differences between the sgRNA targetingMaALS6 andMaALS10 are underlined. (c) Overview of plasmids pETKUL21-PBE-ALS6, pETKUL21-PBE-ALS10 and pETKUL21-PBE-ALS6+10 designed for cytosine base editing. The plant base editor consists of rat cytidine deaminase APOBEC-1, nCas9 (D10A), uracil glycosylase inhibitor (UGI) and nuclear localization signal (NLS), preceded by Zea mays (maize) ubiquitin promoter (ZmUbi) and terminator (T); hptII, hygromycin resistance gene; ntpII, kanamycin resistance gene; ccdB, bacterial cytotoxicity gene;OsU6,Oryza sativa (rice) U6 promoter; LB, left border; RB, right border; PP1, primer pair 1; PP2, primer pair 2. � 2025 The Author(s). New Phytologist� 2025 New Phytologist Foundation. New Phytologist (2025) www.newphytologist.com New Phytologist Research 5 14698137, 0, D ow nloaded from https://nph.onlinelibrary.w iley.com /doi/10.1111/nph.70044 by C ochraneItalia, W iley O nline L ibrary on [15/04/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense Transformation of A. thaliana with any of the unmodified ver- sions of the banana ALS genes did not result in chlorsulfuron resistance. However, editing Pro186 of MaALS6 or Pro181 of MaALS10 in either leucine or phenylalanine did result in trans- genic Arabidopsis resistant to chlorsulfuron (40 lg l�1), indicat- ing that both MaALS6 and MaALS10 function as bona fide ALS genes (Fig. S4). sgRNAs were designed to specifically target the proline residue (Fig. 1b) for both MaALS6 and MaALS10, as C-to-T conversion at the proline residue in both genes has been shown to result in chlorsulfuron resistance in other species (Svitashev et al., 2016; Danilo et al., 2019; Veillet et al., 2019). sgRNAs were first tested in vitro, and both sgRNAs displayed similar efficiency (Fig. S5). We assembled a construct containing the cytosine base editor consisting of a rat cytidine deaminase APOBEC1 fused to an nCas9(D10A) and a uracil glycosylase inhibitor under the control of a maize ubiquitin promoter (Fig. 1c). The constructs pETKUL21-PBE-ALS6, pETKUL21-PBE-ALS10 and pET- KUL21-PBE-ALS6+10 thus included a sgRNA targeting either MaALS6, MaALS10 or both sgRNAs together, respectively (Fig. 1c). An ECS of the banana cultivar Williams was trans- formed with the above-mentioned constructs, and after 6 d, placed on selective medium. After 8 to 12 wk of selection on chlorsulfuron, 104 clumps could be regenerated from ECS trans- formed with pETKUL21-PBE-ALS6, 158 clumps with pETKUL21-PBE-ALS10 and 234 clumps with pETKUL21-PBE- ALS6+10. In total, 496 independent cell clumps grew on chlorsulfuron from c. 2.5 ml settled cell volume (SCV). These individual clumps are expected to be derived from a single cell (Bertsch et al., 2005; Bhatia & Bera, 2015). Differences could be observed in the amount of growing clumps from various cell sus- pensions (P = 0.0074, t-test, df = 34), highlighting the variabil- ity that can occur at any point in the transformation process (Fig. S6). As the construct also harboured a hygromycin resis- tance gene (hptII, Fig. 1c), it was possible to compare selection using hygromycin resistance occurring through the stable T- DNA integration in the banana genome with selection using chlorsulfuron resistance, which would only occur with the C-to- T conversion in at least one of the MaALS genes. For this, an identical amount of SCV was incubated on selective medium containing 50 lg ml�1 hygromycin and 200 mg l�1 timentin. The amount of clumps regenerated on the hygromycin medium was 10- to 30-fold higher in comparison with regeneration on chlorsulfuron (Fig. 2a–d). This difference is expected, as stable transformation events are more likely to directly result in resistant cells in comparison with base edits, which require an extra step post-T-DNA integration and therefore occur more rarely. In comparison with previous work reporting genetic transformation and regeneration of Williams cv (P�erez Hern�andez et al., 2006), we are here reporting regeneration levels that are substantially higher. However, regeneration capacities of different banana cell suspensions have been shown to fluctuate, while the biological features behind this variation have remained elusive (Strosse et al., 2003, 2006). While clumps regenerated on hygromycin are transgenic, they are not necessarily edited and thus resistant to chlorsulfuron. Therefore, a selection of clumps regenerated on hygromycin was transferred to chlorsulfuron-selective plates after 8 to 12 wk, as to make an estimation of the editing rate. For banana transfor- mants with pETKUL21-PBE-ALS6, only nine out of 23 survived, while for banana transformants with pETKUL21-PBE-ALS10, 19 out of 30 survived. Noticeably, 27 out of 34 transformants with pETKUL21-PBE-ALS6+10 could grow on chlorsulfuron. No clumps could regenerate from untransformed cell suspensions on selective medium containing either chlorsulfuron or hygromycin (Fig. 2b,d). Clumps grown on chlorsulfuron-selective plates were further regenerated, and a total of 411 clumps out of 496 regenerated into independent plantlet lines. Each clump mostly resulted in the regeneration of one plant line, with an overall regeneration efficiency of 82.8% (411/496). Lines transformed with plasmids pETKUL21-PBE-ALS6, pETKUL21-PBE-ALS10 and pETKUL21-PBE-ALS6+10 and regenerating on chlorsulfuron were randomly selected in order to profile their edits generated by the base editor at theMaALS6 andMaALS10 target sites. San- ger sequencing indicated that all selected lines harboured at least one edited MaALS allele. Derived allelic variability for selected lines transformed with plasmids pETKUL21-PBE-ALS6, pETKUL21-PBE-ALS10 and pETKUL21-PBE-ALS6+10 was calculated (Figs S7–S9). However, as one chromatogram can sometimes be the result of different mutations, which cannot be distinguished, allelic variability was always interpreted conserva- tively to avoid overestimating mutation rates. One single mutated allele was sufficient to result in chlorsulfuron resistance, as such lines were also retrieved (Fig. 2h). Base edits resulted in nonsy- nonymous mutations replacing Pro186 and Pro181 for MaALS6 and MaALS10, respectively, with amino acids serine, leucine and phenylalanine, as expected (Fig. 2h). Four lines out of 10 targeted with both sgRNAs showed mutations at both target sites. Furthermore, additional mutations were detected in 10 out of 30 lines, including nonsynonymous mutations towards valine, cysteine, arginine, threonine and alanine, as well as frameshift mutations (Figs 2h, S7–S9). Importantly, these nonsynonymous mutations have also been reported to generate chlorsulfuron resis- tance in other plant species (Heap, 2024). Because the sgRNA targeting MaALS6 or MaALS10 only differs by two nucleotides, the sgRNA targeting MaALS6 has a CFD off-target score (Doench et al., 2016) of 0.85 for MaALS10, and the sgRNA tar- getingMaALS10 has a CFD off-target score of 0.79 forMaALS6. As a consequence, two lines out of 10 targeted with pETKUL21- PBE-ALS6 showed edited alleles of MaALS10, and one line out of 10 targeted with pETKUL21-PBE-ALS10 showed edited alleles of MaALS6 (Fig. 2h). In total, 63 mutant alleles were observed across all samples for MaALS6 and MaALS10 com- bined, out of 180. The overall on-target mutation ratio was 27/60 forMaALS6 and 32/60 forMaALS10. Subsequently, chlorsulfuron-resistant lines were screened for the absence of T-DNA. A total of 444 lines were analysed. A PCR-based approach was developed to identify transgene-free lines. Twenty-two lines did not show PCR amplification with a first primer pair targeted at the T-DNA (PP1, Fig. 1c). Lines that appeared negative for the T-DNA PCR1 were subsequently New Phytologist (2025) www.newphytologist.com � 2025 The Author(s). New Phytologist� 2025 New Phytologist Foundation. Research New Phytologist6 14698137, 0, D ow nloaded from https://nph.onlinelibrary.w iley.com /doi/10.1111/nph.70044 by C ochraneItalia, W iley O nline L ibrary on [15/04/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense analysed with a second primer pair (PP2, Fig. 1c). Of these, 17 did not show amplification with both primer pairs (Fig. S10). A full overview of the transgenic state of individual lines can be found in Table S3. Six lines were randomly selected for WGS. No plasmid DNA could be identified in two lines (Fig. 3a,b; Table S4). In another three lines, tiny snippets of reads were pre- sent across the plasmid T-DNA and backbone. Surprisingly, however, this pattern of coverage was also present in both nega- tive controls. Given that this pattern is also in these negative con- trols, it is therefore assumed that this is the result of contamination across samples on either the DNA extraction or library preparation for WGS. Indeed, many of the reads mapping partially to the negative control also mapped partially to coronavirus sequences (Fig. S11). To investigate whether the coverage on these negative controls was similar to those of the edited samples, the coverage by the amount of reads was calculated for each individual base along the plasmid backbone (total of 17 529 and 16 891 bases for pETKUL21-PBE-ALS6 and pETKUL21-PBE-ALS6+10, respectively). Subsequently, the inverted distribution was plotted, showing the amount of bases of the plasmid backbones that were covered by a given amount of reads (Fig. 3). Bases covered by zero reads were omitted from the figure for clarity. Here, five edi- ted lines showed lower or similar distributions compared with Fig. 2 Regeneration of chlorsulfuron-resistant banana calli depends onMaALSmutation. (a) Clumps regenerating on selective medium containing chlorsulfuron and timentin and (b) negative control. (c) Clumps regenerating on selective medium containing hygromycin and timentin and (d) negative control. (e) Chromatogram of clump ALS6+10:21:3 showing the targeted sequence ofMaALS6 and (f)MaALS10. Original sequence can be seen below the chromatogram, and the protospacer adjacent motif site is indicated in red. (g) Edited alleles for line ALS6+10:21-3. Expanded figures can be found in Supporting Information Figs S7–S9. (h) Overview of obtained mutations for clumps targeted with different constructs at their Pro186 and Pro181 sites for MaALS6 andMaALS10, respectively. Desired mutations are indicated in green, and unexpected mutations in grey. Ten transgenic lines per construct were genotyped. AA, amino acid with appropriate three-letter abbreviation; FRM, frameshift. � 2025 The Author(s). New Phytologist� 2025 New Phytologist Foundation. New Phytologist (2025) www.newphytologist.com New Phytologist Research 7 14698137, 0, D ow nloaded from https://nph.onlinelibrary.w iley.com /doi/10.1111/nph.70044 by C ochraneItalia, W iley O nline L ibrary on [15/04/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense the negative controls (Fig. 3). One line (ALS6:21-3) did show substantially higher coverage. Further analysis showed two inde- pendent integration events. A large part of the plasmid backbone of c. 5.5 kb integrated in an intergenic region in Chromosome 5. A second event of c. 1.2 kb integrated in an intron of Mac- ma4_02_g10980, a conserved hypothetical protein on Chromo- some 2, and consisted of a partial hygromycin gene, linked to an inverted fragment of plasmid backbone (Fig. 3a). Both regions Fig. 3 Mapping of whole genome sequencing Illumina reads from transformed banana lines of the cultivar ‘Williams’ to the reference plasmid for (a) pETKUL21-PBE-ALS6 and (b) pETKUL21-PBE-ALS6+10. Low coverage is present among all but two samples (ALS6:9-1 and ALS6+10:21-8). Negative controls (NC), either regenerated from embryogenic cell suspension (ECS), namely NCWilliams ECS, or from in vitro plantlet, namely NCWilliams, show low coverage on both plasmids. ALS6:21-3 contains two independent integration sites of a large part of the plasmid backbone on Chromosome 5 and a fusion of a partial hygromycin gene with additional backbone on Chromosome 2. (c) Line plot showing the number of bases of the plasmid that are covered by a given amount of reads on template pETKUL21-PBE-ALS6 and (d) pETKUL21-PBE-ALS6+10, showing lower coverage for edited samples in comparison with wild-type, except for ALS6:21-3. New Phytologist (2025) www.newphytologist.com � 2025 The Author(s). New Phytologist� 2025 New Phytologist Foundation. Research New Phytologist8 14698137, 0, D ow nloaded from https://nph.onlinelibrary.w iley.com /doi/10.1111/nph.70044 by C ochraneItalia, W iley O nline L ibrary on [15/04/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense were outside the amplified PCR fragments for T-DNA integra- tion, confirming the results of the PCR-based detection of trans- genes in edited banana lines. While the presence of partial T- DNA sequences cannot be excluded in lines identified as transgene-free with the PCR-based approach, the WGS data sug- gest that most of those lines could also be considered transgene- free. Therefore, the transgene-free editing rate could be estimated at c. 3.2% (5/6*17/444 lines) using the above-described method. The target sites in the MaALS genes of these transgene-free lines were investigated for mutations, and the result is visualized in Fig. 4. While most lines show only one edited allele, we were able to regenerate two lines with edits in both genes (ALS6+10:9-13 and ALS6+10:21-10), indicating the possibility of co-editing. In contrast to transgenic samples, no off-target editing could be identified on MaALS6 when targeting MaALS10, or vice versa. The transgene-free gene editing rate was 5.4% (11/204 lines) for pETKUL21-PBE-ALS6+10. Eighteen per cent (2/11 transgene- free lines) harboured mutation at two target sites, indicating sec- ondary mutations can be captured using our editing strategy. The overall efficiency to recover transgene-free gene-edited lines with edits in two distinct target genes is thus estimated at 1.0% (2/204) of all regenerating lines. Lines sent for WGS were screened for off-target mutations at off-target sites with a maxi- mum of four mismatches. However, no C-to-D mutation nor frameshift could be identified within the editing window of the base editor at any of the putative off-target sites, indicating that no off-target effects were present in transgene-free gene-edited lines (Table S5). It should be noted that some off-target sites did already show some level of allelic differences in comparison with the reference genome in the negative controls, as can be seen in Table S5. Discussion As triploid banana contains two bona fide ALS genes, six alleles can be targeted to generate chlorsulfuron resistance. As the tar- geted region is conserved among Musaceae (Fig. S2), the used strategy can easily be expanded to other cultivars for which cell suspensions are already available (Strosse et al., 2006). We observed that transformation with plasmid pETKUL21-PBE- ALS10 appeared to generate a higher proportion of regenerants in comparison with pETKUL21-PBE-ALS6. Moreover, more transgenic clumps selected on hygromycin could survive after transfer to chlorsulfuron, when transformed with pETKUL21- PBE-ALS10. While transcriptomic data indicated that MaALS6 is more actively transcribed (Fig. S3), this might indicate that tar- geted mutations are more easily achieved at the MaALS10 locus in comparison with MaALS6. The sgRNA itself did not seem to effect Cas9 cleavage, as they had similar efficiencies in vitro (Fig. S5). Nevertheless, lines with a single edited allele could be regenerated for both MaALS6 and MaALS10. Most of the trans- genic lines showed at least two MaALS alleles edited. Editing of three alleles, however, appeared to be rare, even in transgenic lines after 12 wk. In wheat, rice and soybean, the level of sulfo- nylurea resistance has been shown to increase with the number of edited ALS alleles (Kawai et al., 2007; R. Zhang et al., 2019; Niu et al., 2024). Most edits observed were clean C-to-T conversion, resulting in nonsynonymous mutations from proline to serine, Fig. 4 Efficient and transgene-free editing of theMaALS genes in banana. (a) Mutations inMaALS genes of transgene-free lines. (b) Regenerating transgene-free plantlets. Amino acids are represented with the appropriate three-letter abbreviation. Bar, 2 cm. � 2025 The Author(s). New Phytologist� 2025 New Phytologist Foundation. New Phytologist (2025) www.newphytologist.com New Phytologist Research 9 14698137, 0, D ow nloaded from https://nph.onlinelibrary.w iley.com /doi/10.1111/nph.70044 by C ochraneItalia, W iley O nline L ibrary on [15/04/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense leucine or phenylalanine. Edited banana lines displaying a pro- line to either valine, cysteine, arginine, threonine or alanine con- version also displayed chlorsulfuron resistance. This is in line with observations made in other dicot and monocot species (Komor et al., 2016; R. Zhang et al., 2019; Heap, 2024). In addition, frameshift mutations were observed at low rates, as the use of the endogenous repair pathway for base excision repair sometimes results in indel mutations due to imprecise nonhomo- logous end-joining (Komor et al., 2016; Ren et al., 2021). Furthermore, one line showed mutations outside the editing win- dow of Bases 3 to 9 distal to the protospacer adjacent motif (PAM) region, resulting in amino acid substitution R187C in MaALS6. This expanded editing window has been reported pre- viously (Lv et al., 2020; Jia et al., 2024), indicating that the edit- ing window for rAPOBEC1-BE3 should be reconsidered to include Base 10. While base editors with expanded editing win- dows and higher mutation efficiencies are readily available in plants (Shimatani et al., 2017; Zong et al., 2018; Jin et al., 2020; Molla et al., 2021; Ren et al., 2021), they also increase the risk of undesired cytosine editing within the editing window, known as bystander editing. Furthermore, although base editors in general show greatly increased specificity in comparison with their Cas9- counterparts (D. Kim et al., 2017), they can also show sgRNA- independent off-target deamination due to the high affinity of the deaminase for single-stranded DNA (Jin et al., 2019; Zuo et al., 2019; Randall et al., 2021; Ren et al., 2021). Notably, measuring such sgRNA-independent off-target effects would require sequencing of edited lines deep enough for unambiguous identification of such mutations. However, the introduction of sgRNA-independent off-target mutations is in the same order of magnitude as mutations arising from somaclonal variation (Ren et al., 2021), and they can be further mitigated through various novel base editor strategies (Jin et al., 2020; Ren et al., 2021; Xiong et al., 2023; He et al., 2024). Because the transient deliv- ery of our base editor system is likely to decrease the incubation time for the introduction of both sgRNA-dependent and sgRNA-independent off-target effects, the level of off-target effects is logically expected to be lower as compared to stable transformation with T-DNA carrying editor cassettes. The stable presence of the Cas9 editing machinery can indeed result in increased off-target effects and mosaicism (Goralogia et al., 2024). WGS and further analysis indicated that five lines out of six were transgene-free. One line (ALS6:21-3) featured two independent transgene insertions, including a plasmid backbone. Integration of the plasmid backbone (Ming et al., 2008) and the T-DNA/backbone conglomerations has both been shown pre- viously (Jupe et al., 2019). While backbone integration is rela- tively common, it further increases the risk for false positives when specifically looking for transgene-free plantlets. Backbone integration is mostly the result of a read-through at the LB sequences, and multiple LB sequences have been shown to reduce unwanted backbone integration (Kuraya et al., 2004; Thole et al., 2007). Moreover, introducing a marker gene just outside the T-DNA near the LB sequence could facilitate early screening for backbone integration (Thole et al., 2007), although WGS would remain essential to verify the absence of any plasmid backbone DNA. While we are certain of a success rate of at least 1.1%, extrapola- tion from 17 samples would allow us to place an expected effi- ciency for transgene-free editing at 3.2% of regenerated lines. Noticeably, the rate of multiple edited alleles was higher in stable transformants as compared to transgene-free edited lines. Neverthe- less, we were able to regenerate at least two transgene-free lines (ALS6+10:9-13 and ALS6+10:21-10) that show mutations at both target sites. Selection on chlorsulfuron facilitates the exclusive regeneration of successfully edited lines. As a result, the probability of recovering lines harbouring additional desired secondary muta- tions is increased in comparison with regeneration on nonselective medium. This strategy can thus be used to enrich for secondary edits at target sites without clear phenotype or selection procedure, in line with recently published co-editing strategies (X. Huang et al., 2023). In banana, this strategy could practically be expanded to perform 18 transformations within one year by one person, requiring the monthly observation and maintenance of c. 2700 clumps, extrapolating to c. 86 transgene-free plantlets, of which 16 would contain mutations at both target sites. Recent publications have reported the aforementioned strategy to enrich for secondary and tertiary mutations in potato, tobacco, tomato and citrus (X. Huang et al., 2023; Jia et al., 2024). The use of markers for the identification of transgenic individuals can further increase the effi- ciency of the strategy, as this would allow the immediate removal of transgenic individuals. Fluorescent reporters have been used extensively in the selection of both transgenic and transgene-free progeny, but are themselves unable to select only mutated transgene-free events (He et al., 2022). Other selection procedures for the removal of transgenic events have been reported, including hypersensitivity reactions to bentazon in rice (Lu et al., 2017) or hygromycin and mannose (Wu et al., 2019; Li et al., 2020). How- ever, these techniques use selection based on susceptibility, with a lower resistance towards bentazon, hygromycin and/or mannose in transgene-free individuals, which unwillingly creates a stressful environment for regenerating plantlets. Therefore, positive selec- tion strategies are usually preferred because they reduce the risk of losing valuable edited lines. While immediate editing through A. tumefaciens in the T0 gen- eration has been shown in multiple crops, including highly hetero- zygous crops and those with long juvenility, this is to the best of our knowledge the first report that shows such transgene-free gene editing in a sterile crop species. As the process of embryogenesis facilitates the regeneration of full plantlets from a single cell, chi- merism is expected to be limited for transgene-free gene-edited lines (Bertsch et al., 2005; Bhatia & Bera, 2015). In contrast to previous studies (Hamada et al., 2018; B�anfalvi et al., 2020; X. Huang et al., 2023; Jia et al., 2024), no chimeric regenerants were observed for transgene-free gene-edited lines based on the investi- gation of the on- and off-target sites. Chimeric regeneration can- not be ruled out for stable transformants, as continuous editing is likely to result in multiple mutations and mosaicism at both on- and off-target sites. Especially in vegetatively propagated crops, chimerism cannot be eliminated through breeding and selection of new planting material requires sequence verification. New Phytologist (2025) www.newphytologist.com � 2025 The Author(s). New Phytologist� 2025 New Phytologist Foundation. Research New Phytologist10 14698137, 0, D ow nloaded from https://nph.onlinelibrary.w iley.com /doi/10.1111/nph.70044 by C ochraneItalia, W iley O nline L ibrary on [15/04/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense With the establishment of a transgene-free gene editing plat- form in banana, it is now possible to develop banana varieties for commercialization with desired edits. The generation of edits in multiple genes in such a co-editing strategy paves the way for effi- cient selection of edited transformants. The combined targeting of MaALS and one or several candidate genes for improved traits will be instrumental in generating and selecting transgene-free bananas harbouring the edits of interest. Importantly, however, the proposed strategy, targeting MaALS to select for desired edits at secondary target sites, is a one-time solution, as selection for continual iterative editing in a single cultivar is not feasible once chlorsulfuron resistance is introduced. Strategies to increase the multiplex capacity during a single transformation event could therefore further increase the strength of the proposed system. Additionally, other target sites could expand the possibilities to introduce further edits in previously improved cultivars. Although not validated in banana, targeted mutations in, for example, acetyl-coenzyme A carboxylase (ACCase) (Liu et al., 2020), 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS ) (Li et al., 2016) and rice tubulin gene OsTubA2 (Liu et al., 2021) have resulted in resistance to aryloxyphenoxypropionate herbi- cides (including haloxyfop), glyphosate and dinitroaniline herbicides, respectively, in several crop species. While banana streak virus sequences, integrated in the B gen- ome of banana cultivars, appear as straightforward target sequences to generate virus-free germplasm for breeding purposes (Tripathi et al., 2019), identification of mutations associated with resistance to stresses, improved nutritional composition or enhanced agronomic performance (Ortiz & Swennen, 2014; Zorrilla-Fontanesi et al., 2020; Tripathi et al., 2024) will help empower the transgene-free editing strategy by providing target genes for the generation of improved banana edited lines. Acknowledgements The authors wish to thank the International Musa Transit Centre (ITC) for the provision of starting material; Sebastien Carpentier of the Alliance of Bioversity and CIAT for fruitful discussion; Edwige Andr�e for her indispensable help in maintaining cell sus- pensions and plant lines; and Barbara Grymonprez for her valu- able aid in conducting the experiments. This work was funded by PhD-fellowship granted by Fonds Wetenschappelijk Onderzoek (FWO) Vlaanderen to SVdB (grant no. 1S22123N) and YN (grant no. 1SHEQ24N). The computing resources and services utilized in this work were provided by the VSC (Flemish Super- computer Center), funded by the Research Foundation – Flan- ders (FWO) and the Flemish Government. Competing interests None declared. Author contributions SVdB, BP and HV designed the experiments and contributed to the writing and editing of the manuscript. SVdB performed experiments and conducted phenotypic characterization and main- tenance of cell lines. YN performed the WGS analysis. SVdB, YN, BP and HV contributed to the interpretation of the results. ORCID Yvan Ngapout https://orcid.org/0000-0002-1719-138X Bart Panis https://orcid.org/0000-0001-6717-947X Senne Van den Broeck https://orcid.org/0000-0002-4493- 5305 Herv�e Vanderschuren https://orcid.org/0000-0003-2102- 9737 Data availability The haplotype-resolved MaALS sequences of Williams are avail- able at Genbank, PQ304390–PQ304395. WGS of transgene- free lines and Williams control is available at NCBI through Bio- project PRJNA1157598. Supporting Information (Figs S1–S11; Tables S1–S5) is provided together with the article. References Ahmad A, Jamil A, Munawar N. 2023. GMOs or non-GMOs? The CRISPR conundrum. Frontiers in Plant Science 14: 1232938. Albright LM, Yanofsky MF, Leroux B, Ma DQ, Nester EW. 1987. Processing of the T-DNA of Agrobacterium tumefaciens generates border nicks and linear, single-stranded T-DNA. Journal of Bacteriology 169: 1046–1055. Alqu�ezar B, Bennici S, Carmona L, Gentile A, Pe~na L. 2022. Generation of transfer-DNA-free base-edited citrus plants. Frontiers in Plant Science 13: 835282. Andersson M, Turesson H, Nicolia A, F€alt AS, Samuelsson M, Hofvander P. 2017. Efficient targeted multiallelic mutagenesis in tetraploid potato (Solanum tuberosum) by transient CRISPR-Cas9 expression in protoplasts. Plant Cell Reports 36: 117–128. Andersson M, Turesson H, Olsson N, F€alt AS, Ohlsson P, Gonzalez MN, Samuelsson M, Hofvander P. 2018. Genome editing in potato via CRISPR- Cas9 ribonucleoprotein delivery. Physiologia Plantarum 164: 378–384. Bae S, Park J, Kim J-S. 2014. Cas-OFFinder: a fast and versatile algorithm that searches for potential off-target sites of Cas9 RNA-guided endonucleases. Bioinformatics 30: 1473–1475. Banakar R, Rai KM, Zhang F. 2022. CRISPR DNA- and RNP-mediated genome editing via Nicotiana benthamiana protoplast transformation and regeneration. In:Methods in molecular biology. Clifton, NJ, USA: Humana Press, 65–82. B�anfalvi Z, Cs�akv�ari E, Vill�anyi V, Kondr�ak M. 2020. Generation of transgene- free PDS mutants in potato by Agrobacterium-mediated transformation. BMC Biotechnology 20: 25. Belhaj K, Chaparro-Garcia A, Kamoun S, Nekrasov V. 2013. Plant genome editing made easy: targeted mutagenesis in model and crop plants using the CRISPR/Cas system. Plant Methods 9: 39. Belser C, Baurens F-C, Noel B, Martin G, Cruaud C, Istace B, Yahiaoui N, Labadie K, H�ribov�a E, Dole�zel J et al. 2021. Telomere-to-telomere gapless chromosomes of banana using nanopore sequencing. Communications Biology 4: 1047. Bertsch C, Kieffer F, Maillot P, Farine S, Butterlin G, Merdinoglu D, Walter B. 2005. Genetic chimerism of Vitis vinifera cv. Chardonnay 96 is maintained through organogenesis but not somatic embryogenesis. BMC Plant Biology 5: 20. Bhatia S, Bera T. 2015. Chapter 6 – Somatic embryogenesis and organogenesis. In: Bhatia S, Sharma K, Dahiya R, Bera TBT-MA of PB in PS, eds.Modern applications of plant biotechnology in pharmaceutical sciences. Boston, MA, USA: Academic Press, 209–230. � 2025 The Author(s). New Phytologist� 2025 New Phytologist Foundation. New Phytologist (2025) www.newphytologist.com New Phytologist Research 11 14698137, 0, D ow nloaded from https://nph.onlinelibrary.w iley.com /doi/10.1111/nph.70044 by C ochraneItalia, W iley O nline L ibrary on [15/04/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://orcid.org/0000-0002-1719-138X https://orcid.org/0000-0002-1719-138X https://orcid.org/0000-0002-1719-138X https://orcid.org/0000-0001-6717-947X https://orcid.org/0000-0001-6717-947X https://orcid.org/0000-0001-6717-947X https://orcid.org/0000-0002-4493-5305 https://orcid.org/0000-0002-4493-5305 https://orcid.org/0000-0002-4493-5305 https://orcid.org/0000-0003-2102-9737 https://orcid.org/0000-0003-2102-9737 https://orcid.org/0000-0003-2102-9737 https://dataview.ncbi.nlm.nih.gov/object/PRJNA1157598 B€orjesson V, Martinez-Monleon A, Fransson S, Kogner P, Johnsen JI, Milosevic J, L�opez MD. 2022. TC-hunter: identification of the insertion site of a transgenic gene within the host genome. BMC Genomics 23: 149. Broad Institute. 2009. Picard toolkit. Cambridge, MA, USA: Broad Institute. [WWW document] URL https://broadinstitute.github.io/picard/ [accessed 20 July 2024]. Camacho C, Coulouris G, Avagyan V, Ma N, Papadopoulos J, Bealer K, Madden TL. 2009. BLAST+: architecture and applications. BMC Bioinformatics 10: 421. Chen L, Li W, Katin-Grazzini L, Ding J, Gu X, Li Y, Gu T, Wang R, Lin X, Deng Z et al. 2018. A method for the production and expedient screening of CRISPR/Cas9-mediated non-transgenic mutant plants. Horticulture Research 5: 13. Concordet J-P, Haeussler M. 2018. CRISPOR: intuitive guide selection for CRISPR/Cas9 genome editing experiments and screens. Nucleic Acids Research 46: W242–W245. Dale J, James A, Paul J-Y, Khanna H, Smith M, Peraza-Echeverria S, Garcia- Bastidas F, Kema G, Waterhouse P, Mengersen K et al. 2017. Transgenic Cavendish bananas with resistance to Fusarium wilt tropical race 4. Nature Communications 8: 1496. Danilo B, Perrot L, Mara K, Botton E, Nogu�e F, Mazier M. 2019. Efficient and transgene-free gene targeting using Agrobacterium-mediated delivery of the CRISPR/Cas9 system in tomato. Plant Cell Reports 38: 459–462. Doench JG, Fusi N, Sullender M, Hegde M, Vaimberg EW, Donovan KF, Smith I, Tothova Z, Wilen C, Orchard R et al. 2016.Optimized sgRNA design to maximize activity and minimize off-target effects of CRISPR-Cas9. Nature Biotechnology 34: 184–191. Dong T, Bi FC, Huang YH, Di HW, Deng GM, Gao HJ, Sheng O, Li CY, Yang QS, Yi GJ et al. 2020.Highly efficient biolistic transformation of embryogenic cell suspensions of banana via a liquid medium selection system. HortScience 55: 703–708. Drenth A, Kema G. 2021. The vulnerability of bananas to globally emerging disease threats. Phytopathology 111: 2146–2161. Droc G, Larivi�ere D, Guignon V, Yahiaoui N, This D, Garsmeur O, Dereeper A, Hamelin C, Argout X, Dufayard J-F et al. 2013. The banana genome hub. Database 2013: bat035. Edgar RC. 2004.MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research 32: 1792–1797. FAO. 2023. Crops and livestock products. [WWW document] URL https://www. fao.org/faostat/en/ [accessed 22 July 2024]. Gao C. 2021. Genome engineering for crop improvement and future agriculture. Cell 184: 1621–1635. Felsenstein J. 1989. PHYLIP-phylogeny inference package (v.3.2). Cladistics 5: 164–166. Goralogia GS, Andreatta IM, Conrad V, Xiong Q, Vining KJ, Strauss SH. 2024. Rare but diverse off-target and somatic mutations found in field and greenhouse grown trees expressing CRISPR/Cas9. Frontiers in Bioengineering and Biotechnology 12: 1412927. Hamada H, Liu Y, Nagira Y, Miki R, Taoka N, Imai R. 2018. Biolistic-delivery- based transient CRISPR/Cas9 expression enables in planta genome editing in wheat. Scientific Reports 8: 14422. He Y, Mudgett M, Zhao Y. 2022. Advances in gene editing without residual transgenes in plants. Plant Physiology 188: 1757–1768. He Y, Han Y, Ma Y, Liu S, Fan T, Liang Y, Tang X, Zheng X, Wu Y, Zhang T et al. 2024. Expanding plant genome editing scope and profiles with CRISPR- FrCas9 systems targeting palindromic TA sites. Plant Biotechnology Journal 22: 2488–2503. Heap I. 2024. The International Herbicide-Resistant Weed Database. [WWW document] URL http://www.weedscience.org/ [accessed 22 July 2024]. Hsu C-T, Chiu C-C, Hsiao P-Y, Lin C-Y, Cheng S, Lin Y-C, Yang Y-L, Wu F- H, Harn H-J, Lin S-Z et al. 2024. Transgene-free CRISPR/Cas9-mediated gene editing through protoplast-to-plant regeneration enhances active compounds in Salvia miltiorrhiza. Plant Biotechnology Journal 22: 1549–1551. Hu C, Liu F, Sheng O, Yang Q, Dou T, Dong T, Li C, Gao H, He W, Liu S et al. 2023. Efficient and transgene-free genome editing in banana using a REG-2 promoter–driven gene-deletion system.Molecular Horticulture 3: 16. Huang H-R, Liu X, Arshad R, Wang X, Li W-M, Zhou Y, Ge X-J. 2023. Telomere-to-telomere haplotype-resolved reference genome reveals subgenome divergence and disease resistance in triploid Cavendish banana. Horticulture Research 10: uhad153. Huang X, Jia H, Xu J, Wang Y, Wen J, Wang N. 2023. Transgene-free genome editing of vegetatively propagated and perennial plant species in the T0 generation via a co-editing strategy. Nature Plants 9: 1591–1597. Huang X, Wang Y, Wang N. 2022. Base editors for citrus gene editing. Frontiers in Genome Editing 4: 852867. Hussain A, Ding X, Alariqi M, Manghwar H, Hui F, Li Y, Cheng J, Wu C, Cao J, Jin S. 2021.Herbicide resistance: another hot agronomic trait for plant genome editing. Plants 10: 621. Illumina Inc. 2023. BCL Convert Standalone v.4.2.4 software release notes. San Diego, CA, USA: Illumina Inc. [WWW document] URL https://www. illumina.com/. Illumina Inc. 2024. Real Time Analysis (RTA3) on Novaseq6000 overview. [WWW document] URL https://knowledge.illumina. com/instrumentation/novaseq-6000/instrumentation-novaseq-6000-reference_ material-list/000002465 [accessed 17 July 2024]. Jaganathan D, Ramasamy K, Sellamuthu G, Jayabalan S, Venkataraman G. 2018. CRISPR for crop improvement: an update review. Frontiers in Plant Science 9: 985. Jao L-E, Wente SR, Chen W. 2013. Efficient multiplex biallelic zebrafish genome editing using a CRISPR nuclease system. Proceedings of the National Academy of Sciences, USA 110: 13904–13909. Jia H, Omar AA, Xu J, Dalmendray J, Wang Y, Feng Y, Wang W, Hu Z, Grosser JW, Wang N. 2024. Generation of transgene-free canker-resistant Citrus sinensis cv. Hamlin in the T0 generation through Cas12a/CBE co- editing. Frontiers in Plant Science 15: 1385768. Jin S, Fei H, Zhu Z, Luo Y, Liu J, Gao S, Zhang F, Chen YH, Wang Y, Gao C. 2020. Rationally designed APOBEC3B cytosine base editors with improved specificity.Molecular Cell 79: 728–740. Jin S, Zong Y, Gao Q, Zhu Z, Wang Y, Qin P, Liang C, Wang D, Qiu J-L, Zhang F et al. 2019. Cytosine, but not adenine, base editors induce genome- wide off-target mutations in rice. Science 364: 292–295. Jupe F, Rivkin AC, Michael TP, Zander M, Motley ST, Sandoval JP, Keith Slotkin R, Chen H, Castanon R, Nery JR et al. 2019. The complex architecture and epigenomic impact of plant T-DNA insertions. PLoS Genetics 15: e1007819. Justine AK, Kaur N, Savita PPK. 2022. Biotechnological interventions in banana: current knowledge and future prospects. Heliyon 8: e11636. Kawai K, Kaku K, Izawa N, Shimizu T, Fukuda A, Tanaka Y. 2007. A novel mutant acetolactate synthase gene from rice cells, which confers resistance to ALS-inhibiting herbicides. Journal of Pesticide Science 32: 89–98. Kennedy B, Thigpen CL. 2020.Many publics around world doubt safety of genetically modified foods. [WWW document] URL https://www.pewresearch. org/short-reads/2020/11/11/many-publics-around-world-doubt-safety-of- genetically-modified-foods/#:~:text=Many%20European%20countries%2C% 20such%20as,being%20unsafe%20rather%20than%20safe [accessed 23 May 2024]. Kim D, Lim K, Kim ST, Yoon SH, Kim K, Ryu SM, Kim JS. 2017. Genome- wide target specificities of CRISPR RNA-guided programmable deaminases. Nature Biotechnology 35: 475–480. Kim H, Kim ST, Ryu J, Kang BC, Kim JS, Kim SG. 2017. CRISPR/Cpf1- mediated DNA-free plant genome editing. Nature Communications 8: 14406. Kleidon J, Brinin A, Paul JY, Harding R, Dale J, Dugdale B. 2020. Production of selectable marker gene-free Cavendish banana (Musa spp.) using a steroid- inducible recombinase platform. Transgenic Research 29: 81–93. Kocsisova Z, Coneva V. 2023. Strategies for delivery of CRISPR/Cas-mediated genome editing to obtain edited plants directly without transgene integration. Frontiers in Genome Editing 5: 1209586. Komor AC, Kim YB, Packer MS, Zuris JA, Liu DR. 2016. Programmable editing of a target base in genomic DNA without double-stranded DNA cleavage. Nature 533: 420–424. Kov�acs G, S�agi L, Jacon G, Arinaitwe G, Busogoro JP, Thiry E, Strosse H, Swennen R, Remy S. 2013. Expression of a rice chitinase gene in transgenic New Phytologist (2025) www.newphytologist.com � 2025 The Author(s). New Phytologist� 2025 New Phytologist Foundation. Research New Phytologist12 14698137, 0, D ow nloaded from https://nph.onlinelibrary.w iley.com /doi/10.1111/nph.70044 by C ochraneItalia, W iley O nline L ibrary on [15/04/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://broadinstitute.github.io/picard/ https://www.fao.org/faostat/en/ https://www.fao.org/faostat/en/ http://www.weedscience.org/ https://www.illumina.com/ https://www.illumina.com/ https://knowledge.illumina.com/instrumentation/novaseq-6000/instrumentation-novaseq-6000-reference_material-list/000002465 https://knowledge.illumina.com/instrumentation/novaseq-6000/instrumentation-novaseq-6000-reference_material-list/000002465 https://knowledge.illumina.com/instrumentation/novaseq-6000/instrumentation-novaseq-6000-reference_material-list/000002465 https://knowledge.illumina.com/instrumentation/novaseq-6000/instrumentation-novaseq-6000-reference_material-list/000002465 https://knowledge.illumina.com/instrumentation/novaseq-6000/instrumentation-novaseq-6000-reference_material-list/000002465 https://knowledge.illumina.com/instrumentation/novaseq-6000/instrumentation-novaseq-6000-reference_material-list/000002465 https://knowledge.illumina.com/instrumentation/novaseq-6000/instrumentation-novaseq-6000-reference_material-list/000002465 https://knowledge.illumina.com/instrumentation/novaseq-6000/instrumentation-novaseq-6000-reference_material-list/000002465 https://knowledge.illumina.com/instrumentation/novaseq-6000/instrumentation-novaseq-6000-reference_material-list/000002465 https://knowledge.illumina.com/instrumentation/novaseq-6000/instrumentation-novaseq-6000-reference_material-list/000002465 https://knowledge.illumina.com/instrumentation/novaseq-6000/instrumentation-novaseq-6000-reference_material-list/000002465 https://knowledge.illumina.com/instrumentation/novaseq-6000/instrumentation-novaseq-6000-reference_material-list/000002465 https://knowledge.illumina.com/instrumentation/novaseq-6000/instrumentation-novaseq-6000-reference_material-list/000002465 https://www.pewresearch.org/short-reads/2020/11/11/many-publics-around-world-doubt-safety-of-genetically-modified-foods/#:~:text=Many%20European%20countries%2C%20such%20as,being%20unsafe%20rather%20than%20safe https://www.pewresearch.org/short-reads/2020/11/11/many-publics-around-world-doubt-safety-of-genetically-modified-foods/#:~:text=Many%20European%20countries%2C%20such%20as,being%20unsafe%20rather%20than%20safe https://www.pewresearch.org/short-reads/2020/11/11/many-publics-around-world-doubt-safety-of-genetically-modified-foods/#:~:text=Many%20European%20countries%2C%20such%20as,being%20unsafe%20rather%20than%20safe https://www.pewresearch.org/short-reads/2020/11/11/many-publics-around-world-doubt-safety-of-genetically-modified-foods/#:~:text=Many%20European%20countries%2C%20such%20as,being%20unsafe%20rather%20than%20safe https://www.pewresearch.org/short-reads/2020/11/11/many-publics-around-world-doubt-safety-of-genetically-modified-foods/#:~:text=Many%20European%20countries%2C%20such%20as,being%20unsafe%20rather%20than%20safe https://www.pewresearch.org/short-reads/2020/11/11/many-publics-around-world-doubt-safety-of-genetically-modified-foods/#:~:text=Many%20European%20countries%2C%20such%20as,being%20unsafe%20rather%20than%20safe banana (‘Gros Michel’, AAA genome group) confers resistance to black leaf streak disease. Transgenic Research 22: 117–130. Kuraya Y, Ohta S, Fukuda M, Hiei Y, Murai N, Hamada K, Ueki J, Imaseki H, Komari T. 2004. Suppression of transfer of non-T-DNA ‘vector backbone’ sequences by multiple left border repeats in vectors for transformation of higher plants mediated by Agrobacterium tumefaciens.Molecular Breeding 14: 309–320. Lampropoulos A, Sutikovic Z, Wenzl C, Maegele I, Lohmann JU, Forner J. 2013. GreenGate – a novel, versatile, and efficient cloning system for plant transgenesis. PLoS ONE 8: e83043. Levi S. 2022. Living standards shape individual attitudes on genetically modified food around the world. Food Quality and Preference 95: 104371. Li B, Sun C, Li J, Gao C. 2024. Targeted genome-modification tools and their advanced applications in crop breeding. Nature Reviews Genetics 25: 603–622. Li H, Durbin R. 2009. Fast and accurate short read alignment with Burrows- Wheeler transform. Bioinformatics 25: 1754–1760. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, Marth G, Abecasis G, Durbin R, 1000 Genome Project Data Processing Subgroup. 2009. The Sequence Alignment/Map format and SAMTOOLS. Bioinformatics 25: 2078–2079. Li J-F, Norville JE, Aach J, McCormack M, Zhang D, Bush J, Church GM, Sheen J. 2013.Multiplex and homologous recombination–mediated genome editing in Arabidopsis and Nicotiana benthamiana using guide RNA and Cas9. Nature Biotechnology 31: 688–691. Li J, Qin R, Zhang Y, Xu S, Liu X, Yang J, Zhang X, Wei P. 2020.Optimizing plant adenine base editor systems by modifying the transgene selection system. Plant Biotechnology Journal 18: 1495–1497. Li J, Meng X, Zong Y, Chen K, Zhang H, Liu J, Li J, Gao C. 2016. Gene replacements and insertions in rice by intron targeting using CRISPR–Cas9. Nature Plants 2: 16139. Li S, Lin D, Zhang Y, Deng M, Chen Y, Lv B, Li B, Lei Y, Wang Y, Zhao L et al. 2022. Genome-edited powdery mildew resistance in wheat without growth penalties. Nature 602: 455–460. Liang Z, Chen K, Li T, Zhang Y, Wang Y, Zhao Q, Liu J, Zhang H, Liu C, Ran Y et al. 2017. Efficient DNA-free genome editing of bread wheat using CRISPR/Cas9 ribonucleoprotein complexes. Nature Communications 8: 14261. Liang Z, Chen K, Zhang Y, Liu J, Yin K, Qiu J, Gao C. 2018. Genome editing of bread wheat using biolistic delivery of CRISPR/Cas9 in vitro transcripts or ribonucleoproteins. Nature Protocols 13: 413–430. Liang Z, Tzfira T. 2013. In vivo formation of double-stranded T-DNA molecules by T-strand priming. Nature Communications 4: 2253. Lin CS, Hsu CT, Yang LH, Lee LY, Fu JY, Cheng QW, Wu FH, Hsiao HCW, Zhang Y, Zhang R et al. 2018. Application of protoplast technology to CRISPR/Cas9 mutagenesis: from single-cell mutation detection to mutant plant regeneration. Plant Biotechnology Journal 16: 1295–1310. Liu L, Kuang Y, Yan F, Li S, Ren B, Gosavi G, Spetz C, Li X, Wang X, Zhou X et al. 2021. Developing a novel artificial rice germplasm for dinitroaniline herbicide resistance by base editing of OsTubA2. Plant Biotechnology Journal 19: 5–7. Liu X, Qin R, Li J, Liao S, Shan T, Xu R, Wu D, Wei P. 2020. A CRISPR- Cas9-mediated domain-specific base-editing screen enables functional assessment of ACCase variants in rice. Plant Biotechnology Journal 18: 1845– 1847. Lowder LG, Zhang D, Baltes NJ, Paul JW, Tang X, Zheng X, Voytas DF, Hsieh TF, Zhang Y, Qi Y. 2015. A CRISPR/Cas9 toolbox for multiplexed plant genome editing and transcriptional regulation. Plant Physiology 169: 971– 985. Lu HP, Liu SM, Xu SL, Chen WY, Zhou X, Tan YY, Huang JZ, Shu QY. 2017. CRISPR-S: an active interference element for a rapid and inexpensive selection of genome-edited, transgene-free rice plants. Plant Biotechnology Journal 15: 1371–1373. Lv X, Qiu K, Tu T, He X, Peng Y, Ye J, Fu J, Deng R, Wang Y, Wu J et al. 2020. Development of a simple and quick method to assess base editing in human cells.Molecular Therapy–Nucleic Acids 20: 580–588. Matsumoto K, Morais LS, Vianna GR, Arag~ao FJL, Rech EL. 2002. Genetic transformation of banana embryogenic cells through particle bombardment using a herbicide resistance gene as selectable marker. Acta Horticulturae 575: 61–67. Ming R, Hou S, Feng Y, Yu Q, Dionne-Laporte A, Saw JH, Senin P, Wang W, Ly BV, Lewis KLT et al. 2008. The draft genome of the transgenic tropical fruit tree papaya (Carica papaya Linnaeus). Nature 452: 991–996. Molla KA, Sretenovic S, Bansal KC, Qi Y. 2021. Precise plant genome editing using base editors and prime editors. Nature Plants 7: 1166–1187. Nekrasov V, Staskawicz B, Weigel D, Jones JDG, Kamoun S. 2013. Targeted mutagenesis in the model plant Nicotiana benthamiana using Cas9 RNA- guided endonuclease. Nature Biotechnology 31: 691–693. Niu Q, Xie H, Cao X, Song M, Wang X, Li S, Pang K, Zhang Y, Zhu J, Zhu J. 2024. Engineering soybean with high levels of herbicide resistance with a Cas12-SF01-based cytosine base editor. Plant Biotechnology Journal 22: 2435– 2437. Ortiz R. 2013. Conventional banana and plantain breeding. Acta Horticulturae 986: 177–194. Ortiz R, Swennen R. 2014. From crossbreeding to biotechnology-facilitated improvement of banana and plantain. Biotechnology Advances 32: 158–169. P�erez Hern�andez JB, Remy S, Swennen R, S�agi L. 2006. Banana (Musa spp.). In: Wang K, ed.Methods in molecular biology. Clifton, NJ, USA: Elsevier, 167–175. Randall LB, Sretenovic S, Wu Y, Yin D, Zhang T, Van EJ, Qi Y. 2021. Genome- and transcriptome-wide off-target analyses of an improved cytosine base editor. Plant Physiology 187: 73–87. Razzaq A, Kaur P, Akhter N, Wani SH, Saleem F. 2021. Next-generation breeding strategies for climate-ready crops. Frontiers in Plant Science 12: 620420. Reed KM, Bargmann BOR. 2021. Protoplast regeneration and its use in new plant breeding technologies. Frontiers in Genome Editing 3: 734951. Remy S, Kov�acs G, Swennen R, Panis B. 2013. Genetically modified bananas: past, present and future. Acta Horticulturae 974: 71–80. Ren Q, Sretenovic S, Liu G, Zhong Z, Wang J, Huang L, Tang X, Guo Y, Liu L, Wu Y et al. 2021. Improved plant cytosine base editors with high editing activity, purity, and specificity. Plant Biotechnology Journal 19: 2052–2068. Robinson JT, Thorvaldsd�ottir H, Winckler W, Guttman M, Lander ES, Getz G, Mesirov JP. 2011. Integrative genomics viewer. Nature Biotechnology 29: 24–26. Rouard M, Droc G, Martin G, Sardos J, Hueber Y, Guignon V, Cenci A, Geigle B, Hibbins MS, Yahiaoui N et al. 2018. Three new genome assemblies support a rapid radiation inMusa acuminata (wild banana). Genome Biology and Evolution 10: 3129–3140. S�agi L, Panis B, Remy S, Schoofs H, De Smet K, Swennen R, Cammue BPA. 1995. Genetic transformation of banana and plantain (Musa spp.) via particle bombardment. Bio/Technology 13: 481–485. Shan Q, Wang Y, Li J, Zhang Y, Chen K, Liang Z, Zhang K, Liu J, Xi JJ, Qiu J- L et al. 2013. Targeted genome modification of crop plants using a CRISPR- Cas system. Nature Biotechnology 31: 686–688. Shimatani Z, Kashojiya S, Takayama M, Terada R, Arazoe T, Ishii H, Teramura H, Yamamoto T, Komatsu H, Miura K et al. 2017. Targeted base editing in rice and tomato using a CRISPR-Cas9 cytidine deaminase fusion. Nature Biotechnology 35: 441–443. Strosse H, Domergue R, Panis B, Escalant J-V, Côte F. 2003. Banana and plantain embryogenic cell suspensions. INIBAP technical guidelines 8. Montpellier, France: The International Network for the Improvement of Banana and Plantain. Strosse H, Schoofs H, Panis B, Andre E, Reyniers K, Swennen R. 2006. Development of embryogenic cell suspensions from shoot meristematic tissue in bananas and plantains (Musa spp.). Plant Science 170: 104–112. Suzuki JY, Tripathi S, Ferm�ın GA, Jan F-J, Hou S, Saw JH, Ackerman CM, Yu Q, Schatz MC, Pitz KY et al. 2008. Characterization of insertion sites in rainbow papaya, the first commercialized transgenic fruit crop. Tropical Plant Biology 1: 293–309. Svitashev S, Schwartz C, Lenderts B, Young JK, Mark Cigan A. 2016. Genome editing in maize directed by CRISPR-Cas9 ribonucleoprotein complexes. Nature Communications 7: 13274. Thole V, Worland B, Snape JW, Vain P. 2007. The pCLEAN dual binary vector system for Agrobacterium-mediated plant transformation. Plant Physiology 145: 1211–1219. Tripathi JN, Ntui VO, Ron M, Muiruri SK, Britt A, Tripathi L. 2019. CRISPR/Cas9 editing of endogenous banana streak virus in the B genome of � 2025 The Author(s). New Phytologist� 2025 New Phytologist Foundation. New Phytologist (2025) www.newphytologist.com New Phytologist Research 13 14698137, 0, D ow nloaded from https://nph.onlinelibrary.w iley.com /doi/10.1111/nph.70044 by C ochraneItalia, W iley O nline L ibrary on [15/04/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense Musa spp. overcomes a major challenge in banana breeding. Communications Biology 2: 46. Tripathi L, Ntui VO, Tripathi JN. 2024. Application of CRISPR/Cas-based gene-editing for developing better banana. Frontiers in Bioengineering and Biotechnology 12: 1395772. Turnbull C, Lillemo M, Hvoslef-Eide TAK. 2021. Global regulation of genetically modified crops amid the gene edited crop boom – a review. Frontiers in Plant Science 12: 630396. Turrell C. 2024. Saving cavendish. Nature Biotechnology 42: 545. Tushemereirwe W, Batte M, Nyine M, Tumuhimbise R, Barekye A, Tendo S, Kubiriba J, Lonenzen J, Swennen R. 2015. Performance of Narita banana hybrids in the preliminary yield trial, Uganda. Kampala, Uganda: National Agriculture Research Organization and International Institute of Tropical Agriculture. Vanderschuren H, Chatukuta P, Weigel D, Mehta D. 2023. A new chance for genome editing in Europe. Nature Biotechnology 41: 1378–1380. Veillet F, Perrot L, Chauvin L, Kermarrec MP, Guyon-Debast A, Chauvin JE, Nogu�e F, Mazier M. 2019. Transgene-free genome editing in tomato and potato plants using Agrobacterium-mediated delivery of a CRISPR/Cas9 cytidine base editor. International Journal of Molecular Sciences 20: 402. Wang Z, Miao H, Liu J, Xu B, Yao X, Xu C, Zhao S, Fang X, Jia C, Wang J et al. 2019.Musa balbisiana genome reveals subgenome evolution and functional divergence. Nature Plants 5: 810–821. Wang Z, Rouard M, Biswas MK, Droc G, Cui D, Roux N, Baurens F-C, Ge X-J, Schwarzacher T, Heslop-Harrison PJS et al. 2022. A chromosome-level reference genome of Ensete glaucum gives insight into diversity and chromosomal and repetitive sequence evolution in theMusaceae. GigaScience 11: giac027. Willems S, Fraiture MA, Deforce D, De Keersmaecker SCJ, De Loose M, Ruttink T, Herman P, Van Nieuwerburgh F, Roosens N. 2016. Statistical framework for detection of genetically modified organisms based on Next Generation Sequencing. Food Chemistry 192: 788–798. Woo JW, Kim J, Il Kwon S, Corval�an C, Cho SW, Kim H, Kim SG, Kim ST, Choe S, Kim JS. 2015. DNA-free genome editing in plants with preassembled CRISPR-Cas9 ribonucleoproteins. Nature Biotechnology 33: 1162–1164. Wu S, Zhu H, Liu J, Yang Q, Shao X, Bi F, Hu C, Huo H, Chen K, Yi G. 2020. Establishment of a PEG-mediated protoplast transformation system based on DNA and CRISPR/Cas9 ribonucleoprotein complexes for banana. BMC Plant Biology 20: 425. Wu T-M, Huang J-Z, Oung H-M, Hsu Y-T, Tsai Y-C, Hong C-Y. 2019. H2O2-based method for rapid detection of transgene-free rice plants from segregating CRISPR/Cas9 genome-edited progenies. International Journal of Molecular Sciences 20: 3885. Xiong X, Liu K, Li Z, Xia FN, Ruan XM, He X, Li JF. 2023. Split complementation of base editors to minimize off-target edits. Nature Plants 9: 1832–1847. Yu Q, Powles SB. 2014. Resistance to AHAS inhibitor herbicides: current understanding. Pest Management Science 70: 1340–1350. Zhang R, Liu J, Chai Z, Chen S, Bai Y, Zong Y, Chen K, Li J, Jiang L, Gao C. 2019. Generation of herbicide tolerance traits and a new selectable marker in wheat using base editing. Nature Plants 5: 480–485. Zhang Y, Liang Z, Zong Y, Wang Y, Liu J, Chen K, Qiu JL, Gao C. 2016. Efficient and transgene-free genome editing in wheat through transient expression of CRISPR/Cas9 DNA or RNA. Nature Communications 7: 12617. Zhao C, Li S, Du C, Gao H, Yang D, Fu G, Cui H. 2022. Establishment of a protoplasts-based transient expression system in banana (Musa spp.). Agronomy 12: 2648. Zong Y, Song Q, Li C, Jin S, Zhang D, Wang Y, Qiu J-L, Gao C. 2018. Efficient C-to-T base editing in plants using a fusion of nCas9 and human APOBEC3A. Nature Biotechnology 36: 950–953. Zong Y, Wang Y, Li C, Zhang R, Chen K, Ran Y, Qiu JL, Wang D, Gao C. 2017. Precise base editing in rice, wheat and maize with a Cas9-cytidine deaminase fusion. Nature Biotechnology 35: 438–440. Zorrilla-Fontanesi Y, Pauwels L, Panis B, Signorelli S, Vanderschuren H, Swennen R. 2020. Strategies to revise agrosystems and breeding to control Fusarium wilt of banana. Nature Food 1: 599–604. Zorrilla-Fontanesi Y, Rouard M, Cenci A, Kissel E, Do H, Dubois E, Nidelet S, Roux N, Swennen R, Carpentier SC. 2016. Differential root transcriptomics in a polyploid non-model crop: the importance of respiration during osmotic stress. Scientific Reports 6: 22583. Zuo E, Sun Y, Wei W, Yuan T, Ying W, Sun H, Yuan L, Steinmetz LM, Li Y, Yang H. 2019. Cytosine base editor generates substantial off-target single- nucleotide variants in mouse embryos. Science 364: 289–292. Supporting Information Additional Supporting Information may be found online in the Supporting Information section at the end of the article. Fig. S1 Natural resistance of banana embryogenic cell suspension of the cultivar ‘Williams’ to chlorsulfuron at different concentra- tions. Fig. S2 Phylogeny of variants of the acetolactate synthase genes in theMusaceae family. Fig. S3 Transcriptional regulation of MaALS6 and MaALS10 in differentMusa species and cultivars. Fig. S4 Chlorsulfuron resistance of transgene Arabidopsis thaliana plants harbouring codon-optimized banana MaALS genes with mutations conferring resistance to chlorsulfuron. Fig. S5 Gel electrophoresis visualizing in vitro cleavage assay of MaALS6 andMaALS10 with designed single-guide RNAs. Fig. S6 Variability between banana embryogenic cell suspension regeneration rates on chlorsulfuron. Fig. S7 Allelic variability after transformation with pETKUL21- PBE-ALS6 in banana. Fig. S8 Allelic variability after transformation with pETKUL21- PBE-ALS10 in banana. Fig. S9 Allelic variability after transformation with pETKUL21- PBE-ALS6+10 in banana. Fig. S10 Gel electrophoresis visualizing amplification of T-DNA in chlorsulfuron-resistant banana clumps. Fig. S11 Reads of negative control (NC) samples NC Williams embryogenic cell suspensions and NC Williams mapping (par- tially) to plasmid pETKUL21-PBE-ALS6+10. Table S1 Genomic location ofMaALS inMusaceae. Table S2 Primers used in this study. Table S3 Overview of T-DNA state of chlorsulfuron-resistant banana clumps. Table S4 Plasmid coverage on banana edited samples sent for whole genome sequencing. New Phytologist (2025) www.newphytologist.com � 2025 The Author(s). New Phytologist� 2025 New Phytologist Foundation. Research New Phytologist14 14698137, 0, D ow nloaded from https://nph.onlinelibrary.w iley.com /doi/10.1111/nph.70044 by C ochraneItalia, W iley O nline L ibrary on [15/04/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense Table S5 Off-target homologous sites with up to four mis- matches to the target sites in theMaALS6 andMaALS10 genes. Please note: Wiley is not responsible for the content or function- ality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office. Disclaimer: The New Phytologist Foundation remains neutral with regard to jurisdictional claims in maps and in any institutional affiliations. � 2025 The Author(s). New Phytologist� 2025 New Phytologist Foundation. New Phytologist (2025) www.newphytologist.com New Phytologist Research 15 14698137, 0, D ow nloaded from https://nph.onlinelibrary.w iley.com /doi/10.1111/nph.70044 by C ochraneItalia, W iley O nline L ibrary on [15/04/2025]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense Outline placeholder Summary Introduction Materials and Methods Plant material Acetolactate synthase sequence analysis Arabidopsis transformation with banana MaALS genes Plasmid generation In vitro cleavage assay Banana transformation and selection on chlorsulfuron DNA purification WGS and data analysis Results Discussion Acknowledgements Competing interests Author contributions References Supporting Information