Engineering herbicide‐resistant sorghum with CRISPR/Cas9‐mediated adenine base editing

An adenine base-editing system was established to precisely modify the sorghum SbALS gene, generating transgene-free mutant plants. These plants exhibit strong herbicide resistance, with no significant differences in agronomic traits, providing valuable germplasm for herbicide-resistance breeding in sorghum. Sorghum (Sorghum bicolor (L.) Moench), the fifth most important cereal crop globally, is used for food, fodder, and biofuel production. Maximizing yield is challenging due to significant losses caused by weed competition, particularly from wild grasses (Pandian et al., 2022). Conventional breeding strategies, including utilization of natural variation and ethyl methanesulfonate (EMS) mutagenesis, have improved herbicide resistance in sorghum (Tang et al., 2025). However, these approaches are often labor-intensive and time-consuming, and their effectiveness is limited by the genetic diversity of cultivated varieties. Clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated nuclease 9 (Cas9)-based gene editing technologies, particularly base editors, offer efficient and precise genome modification. Adenine base editors (ABEs) and cytosine base editors (CBEs) catalyze A·T-to-G·C and C·G-to-T·A transitions, respectively. Glycosylase base editors (GBEs), C-to-G base editors (CGBEs), and A-to-Y base editors (AYBEs) expand the scope of base editing to include transversions (Li et al., 2024). These technologies have generated herbicide resistance in model species such as Arabidopsis (Arabidopsis thaliana) and crops including rice (Oryza sativa), maize (Zea mays), wheat (Triticum aestivum), and soybean (Glycine max) (Zhang et al., 2019; Zhang et al., 2021; Niu et al., 2024; Fu et al., 2025; Nakazato et al., 2025). However, their applicability in sorghum remains unexplored. Acetolactate synthase (ALS) is a key enzyme in branched-chain amino acid biosynthesis. Mutations at the conserved tryptophan 574 (Trp-574, W574) of Arabidopsis ALS or the equivalent position in ALSs from other species confer broad-spectrum resistance to all five classes of ALS-inhibiting herbicides in various species (Figure 1A), including rice, maize, wheat, and Arabidopsis (Pan et al., 2025). To introduce analogous herbicide tolerance in sorghum, we targeted the W545 residue in SbALS. We designed a single guide RNA (sgRNA) for single base editing on the non-coding strand of SbALS to introduce an A-to-G transition, resulting in a Trp-to-Arg substitution (W545R) (Figure 1B). Creation of herbicide-resistant sorghum using a CRISPR-mediated adenine base editing system (A) Partial amino-acid alignment of ALS proteins showing sequence conservation among ALSs from Sorghum bicolor (GenBank: XP_021315526.1), Zea mays (GenBank: NP_001151761.2), Oryza sativa (GenBank: NP_001403551.1), Triticum aestivum (GenBank: XP_044404156.1 for the A subgenome, GenBank: XP_044413937.1 for the B subgenome, and GenBank: XP_044404156.1 for the D subgenome), and Arabidopsis thaliana (GenBank: NP_190425.1). Identical and different amino-acid residues are denoted by dark gray and light gray, respectively. (B) Sequence of the single guide RNA (sgRNA) designed to introduce the W545R mutation in SbALS. The palindrome-adjacent motif (PAM) is in yellow; the target sequence recognized by the sgRNA is in light purple; the targeted codon is underlined, with the target base in red. WT, wild-type. (C) ABE8e vector. NLS, nuclear localization signal; LB, T-DNA left border; RB, T-DNA right border; 35S-CmYLCV-SbU6, composite promoter consisting of 35S, CmYLCV, and native SbU6 promoters. (D) Table summarizing the Agrobacterium tumefaciens-mediated genetic transformation of P184, a conventional grain sorghum variety. (E) Sanger chromatograms showing the target site in WT (left) and regenerant T0 plant #5 (right). The PAM is in yellow; the target is in light purple; the targeted base is in red. (F) Representative photographs of WT and SbALSW545R/SbALS plants at 8, 16, or 24 d after spraying with 200 or 400 mg/L imazethapyr (or water only as control [0 mg/L]) at the 4–5 leaf stage in the greenhouse. DAS, the day after spraying. (G) Representative photographs of WT and SbALSW545R/SbALS plants at 16 d after spraying with 200 mg/L imazethapyr or water only (0 mg/L) in the field. (H) Plant height, tiller number, stem diameter, panicle length, single panicle weight, and thousand-grain weight of WT and SbALSW545R/SbALS plants. Data are means ± standard error (n = 10). P-values were calculated by Student's t-test. ns, not significant. We developed a sorghum-specific ABE8e vector by cloning a codon-optimized TadA-8e in-frame and upstream of the sequence encoding the nickase Cas9 (nCas9; Cas9D10A) variant. To enhance base-editing efficiency, we used the native SbU6 promoter in combination with a composite cauliflower mosaic virus (CaMV) 35S promoter and the Cestrum yellow leaf curling virus (CmYLCV) promoter to drive tRNA-sgRNA-HDV expression (Jiang et al., 2020; Figure 1C). We tested the editing efficiency of this construct by transfecting it into sorghum protoplasts; deep sequencing of PCR products encompassing the targeted site using the Hi-TOM platform showed evidence of the expected A-to-G change, with an average editing efficiency of 11.6% based on three independent experiments (10.3%, 11.9%, 12.7%) (Figure S1). We introduced this ABE construct into the “P184” cultivar using the Agrobacterium (Agrobacterium tumefaciens)-mediated immature embryo method, generating 17 regenerated plants from 217 immature embryos. Sanger sequence confirmed that 12 regenerated plants carried transgenic elements, among which four were heterozygous for the edited SbALSW545R allele, corresponding to an editing efficiency of 33.3% at the desired target (Figures 1D, E, S2). No off-target editing was detected at the five predicted sites (Figure S3; Table S1). Future genome-wide analyses will be needed to fully evaluate potential off-target effects in sorghum. To obtain stable, transgene-free homozygous mutant plants, we examined the progeny of the four SbALSW545R/SbALS plants for the presence of transgene elements and the mutation. Segregation of the transgene elements from the T0 to the T1 generation followed Mendelian inheritance, indicating that the parental plants carried a single copy of the transgene (Table S2). However, none of the T1 progeny were homozygous for the SbALSW545R allele, regardless of their transgene status. Further genotyping of the progeny from transgene-free T1 plants #5-5, #5-15, and #5-27 heterozygous for SbALSW545R revealed 2:1 segregation between heterozygotes and plants with the wild-type genotype, with no plants homozygous for the SbALSW545R allele recovered (Table S2). These findings suggest that the ABE-mediated SbALSW545R edits can be stably transmitted to the next generation. However, the failure to recover homozygous SbALSW545 indicates that the W545R substitution may impair essential function when present as the sole ALS form and therefore be more suitable for deployment in the heterozygous state. Before assessing the herbicide resistance of SbALSW545R/SbALS plants, we first determined the lethal concentration of imazethapyr for P184 plants by spraying 4–5-leaf-stage plants with a range of imazethapyr concentrations. A concentration of 50 mg/L imazethapyr caused complete lethality and was therefore defined as 1× (Figure S4). We then sprayed WT and transgene-free SbALSW545R/SbALS plants with 200 mg/L (4×) or 400 mg/L (8×) imazethapyr. Eight days after herbicide treatment, WT plants were chlorotic, and all WT plants had died by day 24. By contrast, the SbALSW545R/SbALS plants had minimal growth inhibition at 200 mg/L and only mild growth suppression at 400 mg/L (Figure 1F). Thus, SbALSW545R/SbALS plants exhibit greater tolerance to imazethapyr than WT plants. We further evaluated the herbicide resistance of these SbALSW545R/SbALS plants under field conditions. One month after sowing, we sprayed all plants with 200 mg/L imazethapyr (4×) or water only. At 16 d post-treatment, all imazethapyr-treated WT plants had died, whereas SbALSW545R/SbALS plants sprayed with the herbicide displayed vigor similar to untreated WT plants (Figure 1G). At maturity, there were no significant differences in agronomic traits, such as plant height, stem diameter, tiller number, panicle length, panicle weight, and thousand-grain weight, between herbicide-treated SbALSW545R/SbALS and untreated WT plants (Figure 1H). Consistent with the observed herbicide tolerance, the mutant ALS protein retained higher enzymatic activity than the wild-type protein in the presence of herbicides from five ALS-inhibitor classes (Figure S5). In conclusion, this work demonstrates that an ABE gene editing system can precisely replace an intended single nucleotide in the grain crop sorghum. By targeted modification of SbALS, we created sorghum germplasm more resistant to imazethapyr, providing a resource for breeding herbicide-resistant sorghum. As base-editing and prime-editing technologies continue to evolve, they will accelerate sorghum trait improvements, including herbicide resistance, yield, and quality. This work was supported by the National Natural Science Foundation of China (32241041) and the Agricultural Science, Technology Innovation Program of the Chinese Academy of Agricultural Sciences (01-ICS-22), and the Scientific Research Team of Zhongyuan Research Center, Chinese Academy of Agricultural Sciences (CAAS-ZRC-ZYZX20230203). The authors declare no conflicts of interest. B.D., Q.Q., and J.Z. designed the experiments. J.Z., R.L., Z.W., X.S., L.S., X.F., F.L., J.Z., and G.L. performed the experiments. J.Z. wrote the manuscript. B.D. and J.Z revised the manuscript. All authors read and approved the final manuscript. Additional Supporting Information may be found online in the supporting information tab for this article: http://onlinelibrary.wiley.com/doi/10.1111/jipb.70298/suppinfo Figure S1. Precise efficiency of the SbALSW545R target editing in sorghum protoplasts. Figure S2. Genotypes of SbALS for the ABE target site in T0 and T1 regenerant plants. Figure S3. Representative Sanger sequencing chromatograms of off-targets. Figure S4. The minimal lethal concentration in wild-type plants with imazethapyr. Figure S5. Dose-response of ALS protein activity in five classes of ALS inhibitor herbicides. Table S1. Analysis of potential off-target sites for the ABE construct designed to introduce the SbALSW545R mutation. Table S2. Transmission and segregation of the SbALSW545R mutation in T1 and T2 generations. Tables S3. Primers used in this study. Method S1. Materials and methods used in this study Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.