Introduction

CRISPR-Cas12a (formerly Cpf1) is a class II type V endonuclease that prefers a thymine-rich protospacer adjacent motif (PAM)1, and is the second most commonly used CRISPR system for genome editing2,3,4. Cas12a has shown higher targeting specificity than Cas9 in mammalian cells and plants5,6,7. Unlike Cas9, Cas12a generates staggered ends and larger deletions, making it a suitable nuclease for gene knockout. Moreover, Cas12a only requires a short CRISPR RNA (crRNA) for each target and possesses RNase activity for crRNA array processing, making it an excellent platform for multiplexed editing8,9. Cas12a has been used to generate targeted mutations as well as achieve transcriptional regulation in microorganisms10,11,12. Cas12a can also efficiently generate edits in some important industrial Streptomyces strains that cannot be edited using SpCas9 due to toxicity12. In mammalian systems, Cas12a has been widely used and engineered for genetic manipulation1,13,14,15. Three Cas12a orthologs, LbCas12a, AsCas12a, and FnCas12a, have been used in plants. LbCas12a is most popular due to its high editing activity in rice9,16,17,18,19,20,21,22, maize23,24, Arabidopsis24,25,26, tomato25, Nicotiana benthamiana25, lettuce27, cotton28, and citrus29. FnCas12a also mediates efficient genome editing in plants9,17,20,30. While most Cas12a studies in plants pursued targeted mutagenesis by non-homologous end joining (NHEJ) DNA repair, precise genome editing based on homology-directed repair (HDR) has also been demonstrated20,22,31,32,9,25,66.

Mutation analysis by RFLP and SSCP

To analyze mutation results using the restriction fragment length polymorphism (RFLP) method, the targeted genomic regions were amplified, and the PCR products were digested with restriction enzymes whose cutting sites are overlap** with the expected editing sites. Digested products were visualized on 2% TAE agarose gels. Mutation frequencies were quantified based on band intensity using Image Lab™ Software (Bio-Rad Laboratories) and ImageJ (https://imagej.nih.gov/ij/) for protoplast assays. For stable transgenic rice lines, any samples with undigested bands were considered edited, while any samples without detectable digested bands were considered biallelically edited in the RFLP analysis. To analyze mutation results in transgenic rice plants using the single-strand conformational polymorphism (SSCP) method67, amplicons were denatured for 5 min at 95 °C and immediately put on ice to minimize self-annealing. Denatured PCR amplicons were electrophoresed on 15 % non-denaturing polyacrylamide gels at 45 mA, 120–200 V. After 6 h, polyacrylamide gels were stained using argentation.

Sanger sequencing and deep sequencing

PCR amplicons from stable transgenic rice were subjected to Sanger sequencing. DNA sequences were decoded using DSDecodeM68. PCR amplicons with sequencing barcodes generated from protoplast assay were first submitted for quality check, followed by sequencing using an Illumina HiseqXten-PE150 system. Sequencing data were then analyzed by CRISPRMatch69 for editing frequencies and profiles. For off-target analysis, amplicons from Line 10 (multiplexed genome editing by Mb2Cas12a targeting four sites) and Line 1 (multiplexed genome editing by Mb2Cas12a targeting six sites for quantitative traits) were pooled together, while amplicons from Line 11 (multiplexed genome editing by Mb2Cas12a targeting four sites) and Line 2 (multiplexed genome editing by Mb2Cas12a targeting six sites for quantitative traits) were pooled together. Potential off-target sites were sequenced using the Illumina HiSeq2500 system and analyzed using CRISPResso270. To assess the editing results for multiplexed genome editing at 4 sites and 16 sites using system B, PCR amplicons were barcoded using the Hi-TOM46 primers and pooled into two and four samples for Illumina HiSeq2500, respectively. Data were analyzed using the Hi-TOM online tool (https://doi.org/www.hi-tom.net/hi-tom/) with a 12.5% filter threshold.

Arabidopsis stable transformation

Arabidopsis thaliana wild type plants Col-0 were transformed with Agrobacterium tumefaciens GV3101 using the flora dip method71. T1 generation seeds were collected and sterilized with 50% bleach and 0.05% Tween. After 3 days of vernalization at 4 °C, seeds were plated on ½ MS medium supplemented with 15 mg l−1 hygromycin. After a week, hygromycin resistant plants were transferred to ½ MS medium to recover for another week. Individual plants were treated as single transgenic lines. Plants transformed with a GUS gene and a hygromycin resistance gene were used as controls. Leaf tissue was collected from each line along with control plants for qRT-PCR analysis. Seeds collected from these lines were treated using the same method to obtain the T2 generation plants. Three plants of each line were used for qRT-PCR analysis.

Gene expression analysis using qRT-PCR

Total RNA from rice and Arabidopsis leaf tissue or rice protoplast was extracted using TRIzol™ Reagent (Thermo Fisher Scientific) following the manufacturer’s instructions with slight modifications. DNA was then removed with DNase I (New England BioLabs) and the complementary DNA (cDNA) was synthesized using the SuperScriptTM III First-Strand Synthesis System (Thermo Fisher Scientific). Applied Biosystems™ SYBR™ Green PCR Master Mix (Thermo Fisher Scientific) was used for qRT-PCR. Three technical replicates were carried out for each sample using a CFX96 Touch™ Real-Time PCR Detection System (Bio-Rad Laboratories). AtEf1α was used as the reference gene for Arabidopsis and OsTubulin was used as the reference gene for rice. Protoplasts transformed with the T-DNA vector containing LbCas12a without crRNAs were used as controls for rice. Relative expression level of each gene was calculated using the 2−ΔΔCt method.

Genome-wide PAM analysis

The rice cultivar Nipponbare genome sequence (Oryza_sativa_nipponbare_v7.0_all.con) was obtained from the Rice Genome Annotation Project (rice.plantbiology.msu.edu/pub/data/Eukaryotic_Projects/o_sativa/annotation_dbs/pseudomolecules/). PAM sequence and its reverse complementary sequence were used to search the FASTA files using a regular expression search in Perl (version 5.26).

Statistical analysis

Pearson’s χ2 test was used to show whether there is a significant difference between the expected frequencies and the observed frequencies of edits in transgenic rice T1 populations when α = 0.05. Student’s t-test was used for pairwise comparison. One asterisk (p < 0.05) and two asterisks (p < 0.01) indicate significant differences between two treatments. Tukey’s Honest Significant Difference (HSD) test was used for multiple comparisons. Treatments with the same letter are not significantly different when α = 0.05.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.