Prime editors, which are CRISPR-Cas9 nickase (H840A)reverse transcriptase fusions programmed with prime editing guide RNAs (pegRNAs), can edit bases in mammalian cells without donor DNA or double-strand breaks. We adapted prime editors for use in plants through codon, promoter, and editing-condition optimization. The resulting suite of plant prime editors enable point mutations, insertions and deletions in rice and wheat protoplasts. Regenerated prime-edited rice plants were obtained at frequencies of up to 21.8%. Introduction of genome modifications such as substitutions, insertions, and deletions that improve agronomic traits can accelerate crop improvement and breeding 1,2. In plants, nuclease-initiated homology-directed repair (HDR) is limited by low efficiency and the difficulty of DNA template delivery 3-6. Cytosine and adenine base editors (CBEs and ABEs) install C•G-to-T•A and A•T-to-G•C transitions 7-9 , and have been successfully used in plants 4. However, base editors are unable to install transversions, insertions, or deletions 7-10. Prime editing uses engineered Cas9 nickase-reverse transcriptase (RT) fusion proteins paired with a pegRNA that encodes the desired edit 11. The RT domain uses a nicked genomic DNA strand as a primer for the synthesis of an edited DNA flap templated by an extension on the pegRNA. Subsequent DNA repair incorporates the edited flap, permanently installing the programmed edit 11. To optimize prime editing for plants, we first compared three plant prime editor systems (PPEs): PPE2, PPE3, and PPE3b 11 (Fig. 1a). PPE2 consists of a nCas9(H840A) fused to an engineered M-MLV RT, and a pegRNA composed of a primer binding site (PBS) and an RT template 11. PPE3 adds an additional nicking single guide RNA (sgRNA) to cleave the non-edited strand, which facilitates favorable DNA repair. In PPE3b, this nicking sgRNA targets the edited sequence, thereby preventing nicking of the non-edited strand until after editing occurs, resulting in fewer indels in mammalian cells 11. We codon-optimized PPE genes for cereal plants and expressed them using the maize Ubiquitin-1 (Ubi-1) promoter (Fig. 1b). We used the OsU3 (or TaU6) and TaU3 promoters to drive pegRNA and nicking sgRNA transcription, respectively. To test whether other RTs support prime editing, we replaced the engineered M-MLV RT with either the CaMV RT (RT-CaMV) from cauliflower mosaic virus 12 or a retron-derived RT (RT-retron) from E. coli BL21 (ref. 13) (Fig. 1b). We first used our previously described 14 rice protoplast reporter system to test the PPE system for blue fluorescent protein (BFP) to green fluorescent protein (GFP) conversion, which requires changing codon 66 from CAC (histidine) to TAC (tyrosine)

Cytosine and adenine base editors (CBEs and ABEs) are promising new tools for achieving the precise genetic changes required for disease treatment and trait improvement. However, genome-wide and unbiased analyses of their off-target effects in vivo are still lacking. Our whole genome sequencing (WGS) analysis of rice plants treated with BE3, high-fidelity BE3 (HF1-BE3), or ABE revealed that BE3 and HF1-BE3, but not ABE, induce substantial genome-wide off-target mutations, which are mostly the C→T type of single nucleotide variants (SNVs) and appear to be enriched in genic regions. Notably, treatment of rice with BE3 or HF1-BE3 in the absence of single-guide RNA also results in the rise of genome-wide SNVs. Thus, the base editing unit of BE3 or HF1-BE3 needs to be optimized in order to attain high fidelity.

No abstract