Expanding the scope of plant genome engineering with Cas12a orthologs and highly multiplexable editing systems

Abstract

CRISPR-Cas12a is a promising genome editing system for targeting AT-rich genomic regions. Comprehensive genome engineering requires simultaneous targeting of multiple genes at defined locations. Here, to expand the targeting scope of Cas12a, we screen nine Cas12a orthologs that have not been demonstrated in plants, and identify six, ErCas12a, Lb5Cas12a, BsCas12a, Mb2Cas12a, TsCas12a and MbCas12a, that possess high editing activity in rice. Among them, Mb2Cas12a stands out with high editing efficiency and tolerance to low temperature. An engineered Mb2Cas12a-RVRR variant enables editing with more relaxed PAM requirements in rice, yielding two times higher genome coverage than the wild type SpCas9. To enable large-scale genome engineering, we compare 12 multiplexed Cas12a systems and identify a potent system that exhibits nearly 100% biallelic editing efficiency with the ability to target as many as 16 sites in rice. This is the highest level of multiplex edits in plants to date using Cas12a. Two compact single transcript unit CRISPR-Cas12a interference systems are also developed for multi-gene repression in rice and Arabidopsis. This study greatly expands the targeting scope of Cas12a for crop genome engineering.

Fig. 1. Targeted mutagenesis by Cas12a orthologs at canonical TTTV PAM sites in rice.

a A schematic diagram of the Cas12a expression system for single site targeting. pZmUbi maize ubiquitin promoter, NLS nuclear localization signal, tNOS nopaline synthase (NOS) terminator, HH hammerhead ribozyme, HDV hepatitis delta virus ribozyme. b Targeted mutagenesis efficiencies (percentage) of ten Cas12a orthologs at two TTTV PAM sites in rice protoplasts. WT, protoplasts transformed with water. c Targeting specificity of four Cas12a orthologs measured with mismatched crRNAs at the OsEPFL9-TTTG site in rice protoplasts. Mismatched nucleotides in crRNAs are highlighted in green. d Protospacer length requirements of four Cas12a orthologs at the OsEPFL9-TTTG site in rice protoplasts. e Summary of editing and biallelic editing efficiencies of Cas12a orthologs at two TTTV PAM sites in transgenic rice T₀ lines. f Editing analysis of transgenic rice T₁ lines. χ² test results indicate that T₁ populations from each tested T₀ line segregated at a 1:2:1 ratio when α = 0.05. Data in “b–d” are presented as mean values ± SEM. n = 3 biologically independent samples. Source data are provided as a Source Data file.

Fig. 2. Mb2Cas12a and engineered Cas12a variants broaden the genome editing scope in rice.

a Targeted mutagenesis efficiencies (percentage) of ten Cas12a orthologs at two VTTV PAM sites in rice protoplasts. WT, protoplasts transformed with water. b Targeted mutagenesis efficiencies (percentage) of Mb2Cas12a at 18 VTTV PAM sites in rice protoplasts. c Summary of editing and biallelic editing efficiencies by Mb2Cas12a at two VTTV PAM sites in transgenic rice T₀ lines. d Comparison of editing efficiencies (percentage) of five Cas12a RVR variants at six TATV PAM sites in rice protoplasts. Treatments with the same letter are not significantly different when α = 0.05 by Tukey’s Honest Significant Difference test (two-sided). e A box plot showing targeted mutagenesis efficiencies (percentage) of Mb2Cas12a-RVRR variant at 51 canonical and altered PAM sites in rice protoplasts. V = A, C, and G. Y = T and C. S = C and G. R = A and G. Each data point represents the mean of three biological replicates (n = 3). The middle line in each box denotes the median. The upper and lower bounds of each box denote the first quartile (25%) and the third quartile (75%), respectively. The top and bottom of whiskers denote the maxima and minima, respectively. f Genome-wide analysis of targetable PAM sites by Mb2Cas12a-RVRR variant in rice (Oryza sativa). g and h, Insertion and deletion (Indel) frequencies (percentage) of four Cas12a orthologs at two TTTV PAM sites at 32 and 22 °C in rice protoplasts. One asterisk (p < 0.05) and two asterisks (p < 0.01) indicate significant differences between two treatments using two-sided Student’s t-test. p = 0.000583 (Er vs. Mb2); p = 0.000456 (Lb5 vs. Mb2); p = 0.012092 (Bs vs. Mb2). i Indel frequencies (percentage) of Mb2Cas12a at two VTTV PAM sites at 32 and 22 °C in rice protoplasts. Data in a, b, d, g–i are presented as mean values ± SEM. n = 3 biologically independent samples. Source data are provided as a Source Data file.

Fig. 3. Comparison of 12 multiplex Cas12a genome editing systems in rice.

a Schematics of 12 multiplexed Cas12a systems, which are classified into six different strategies. The schematics are based on multiplexing four crRNAs. pZmUbi maize ubiquitin promoter, NLS nuclear localization signal, tNOS nopaline synthase (NOS) terminator, pU6 rice U6 promoter, pU3 rice U3 promoter, HH hammerhead ribozyme, HDV hepatitis delta virus ribozyme, pT polyT, pA polyA, DR direct repeat. b Comparison of ten multiplexed Cas12a editing systems (A–J) in stable transgenic rice lines, by using LbCas12a to target OsDEP1 and OsROC5 with four crRNAs. Approximately three independent T₀ lines were genotyped at each target site for wild type (denoted as an empty rectangle), monoallelic mutant (denoted as a half-filled rectangle) and biallelic mutant (denoted as a fully filled rectangle). c An example of a quadruple mutant showing loss-of-function phenotypes for OsPDS and OsROC5, compared to the wild type (WT). Additional quadruple mutants with similar phenotypes were identified. Size bar, 4 cm. d Comparison of seven multiplexed editing systems (B, D, G, H, M, I, and L) in stable transgenic rice lines, using LbCas12a to target four genes (OsPDS, OsDEP1, OsROC5, and OsmiR528) with four crRNAs. The number of transgenic T₀ plants assayed are also indicated by “n”. e Further analysis of the data in “d” for editing efficiencies of each multiplexed LbCas12a system when editing certain defined numbers of genes (4, 3, 2, and 1) simultaneously. f Further analysis of the data in “d” for biallelic editing efficiencies of each multiplexed LbCas12a system when editing certain defined numbers of genes (4, 3, 2, and 1) simultaneously. g Multiplexed editing of four genes with system B by Mb2Cas12a. Genotypes at each target site in 11 T₀ lines are shown with wild type denoted as an empty rectangle, monoallelic mutant denoted as a half-filled rectangle and biallelic mutant denoted as a fully filled rectangle. Source data are provided as a Source Data file.

Fig. 4. Highly efficient large-scale biallelic genome editing in rice.

a Simultaneous editing of six sites by Mb2Cas12a for multiplexed engineering of quantitative traits in rice. Genotypes of four independent T₀ lines are shown, indicating all the target sites are nearly edited biallelically. The PAM sites are highlighted in red and the target sequences are highlighted in blue. The TATA boxes are painted in red. The TAL effector binding sites in the promoters of the SWEET genes are indicated by solid black lines above the sequences. Exons of GS3 and GW2 are underlined. b Schematics of simultaneous editing by LbCas12a at 16 different target sites (T1–T16) across nine rice chromosomes (chr). c Genotypes of 21 tested T₀ lines at 16 target sites, with wild type denoted as empty rectangles, monoallelic mutants denoted as half-filled rectangles, biallelic mutants denoted as fully filled rectangles, and chimeras denoted as doted rectangles. d Genotypes of T₀ Line 2 and Line 21 at 16 target sites, where 14 sites were biallelically edited in both lines.