Sequencing results of the rBE14-induced OsWRKY45 mutations in T0 transgenic rice callus lines. (I) Summary of nucleotide changes in the editing window of the endogenous OsWRKY45 gene in T0 transgenic rice callus lines. (J) Summary of base editing efficiencies of rBE14 with different DNA contexts at five genomic sites. (K) Visualizing GFP in transgenic rice callus under a handheld UV lamp and confocal microscope. For (B), (E), (G), (H), and (I), The PAM sequence, the candidate bases in the putative editing window, the detected nucleotide changes/the corresponding amino acids and the sgRNA sequence are highlighted in green, red, blue and bold, respectively. For (C) and (F), the nucleotide changes are underlined in the sequencing chromatograms.

Recently developed CRISPR-mediated base editors, which enable the generation of numerous nucleotide changes in target genomic regions, have been widely adopted for gene correction and generation of crop germplasms containing important gain-of-function genetic variations. However, to engineer target genes with unknown functional SNPs remains challenging. To address this issue, we present here a base-editing-mediated gene evolution (BEMGE) method, employing both Cas9n-based cytosine and adenine base editors as well as a single-guide RNA (sgRNA) library tiling the full-length coding region, for developing novel rice germplasms with mutations in any endogenous gene. To this end, OsALS1 was artificially evolved in rice cells using BEMGE through both Agrobacterium-mediated and particle-bombardment-mediated transformation. Four different types of amino acid substitutions in the evolved OsALS1, derived from two sites that have never been targeted by natural or human selection during rice domestication, were identified, conferring varying levels of tolerance to the herbicide bispyribac-sodium. Furthermore, the P171F substitution identified in a strong OsALS1 allele was quickly introduced into the commercial rice cultivar Nangeng 46 through precise base editing with the corresponding base editor and sgRNA. Collectively, these data indicate great potential of BEMGE in creating important genetic variants of target genes for crop improvement.

Many phytopathogens secrete cell wall degradation enzymes (CWDEs) to damage host cells and facilitate colonization. As the major components of the plant cell wall, cellulose and hemicellulose are the targets of CWDEs. Damaged plant cells often release damage-associated molecular patterns (DAMPs) to trigger plant immune responses. Here, we establish that the fungal pathogen Magnaporthe oryzae secretes the endoglucanases MoCel12A and MoCel12B during infection of rice (Oryza sativa). These endoglucanases target hemicellulose of the rice cell wall and release two specific oligosaccharides, namely the trisaccharide 31-β-D-Cellobiosyl-glucose and the tetrasaccharide 31-β-D-Cellotriosyl-glucose. 31-β-D-Cellobiosyl-glucose and 31-β-D-Cellotriosyl-glucose bind the immune receptor OsCERK1 but not the chitin binding protein OsCEBiP. However, they induce the dimerization of OsCERK1 and OsCEBiP. In addition, these Poaceae cell wall-specific oligosaccharides trigger a burst of reactive oxygen species (ROS) that is largely compromised in oscerk1 and oscebip mutants. We conclude that 31-β-D-Cellobiosyl-glucose and 31-β-D-Cellotriosyl-glucose are specific DAMPs released from the hemicellulose of rice cell wall, which are perceived by an OsCERK1 and OsCEBiP immune complex during M. oryzae infection in rice.

The target site of pi-d2 gene in rice. (D) Representative Sanger sequencing chromatogram of the rBE5-edited pi-d2 alleles with the target G changed in T0 transgenic rice line #3. (E) Efficiency of rBE5-versus rBE3-mediated gene correction of pi-d2 in T0 transgenic rice lines. (F) The target site of OsFLS2 gene in rice. (G) Representative Sanger sequencing chromatogram of the rBE5-edited OsFLS2 alleles with anticipated C > A conversion in T0 transgenic rice line #21. (H) Efficiency of the rBE5 system in base editing OsFLS2 in T0 transgenic rice lines. (I) Sequencing results of the rBE5-induced OsFLS2 mutations in T0 transgenic rice lines. (J) Construct of the rBE9 system used to induce nucleotide changes in rice transgenic plants. (K) Efficiencies and ratios of allelic mutations caused by the rBE3 and rBE9 systems. (L) Base editing efficiencies of the rBE3 and rBE9 systems at the target C in different sequence context. For (C, F, and I), the PAM sequences, the putative target bases in the activity window, and the detected nucleotide/corresponding amino acids are highlighted in green, red, and blue, respectively. For (D and G), nucleotide mutations are underlined in the sequencing chromatograms.

CRISPR technologies enabling precise genome manipulation are valuable for gene function studies and molecular crop breeding. However, the requirement of a protospacer adjacent motif (PAM), such as NGG and TTN, for Cas protein recognition restricts the selection of targetable genomic loci in practical applications of CRISPR technologies. Recently Cas9-NG, which recognizes a minimal NG PAM, was reported to expand the targeting space of genome editing in human cells, but it remains unclear whether this Cas9 variant can be used in plants. In this study, we evaluated the nuclease activity of Cas9-NG toward various NGN PAMs by targeting endogenous genes in transgenic rice. We found that Cas9-NG edits all NGG, NGA, NGT, and NGC sites with impaired activity, while the gene-edited plants were dominated by monoallelic mutations. Cas9-NG-engineered base editors were then developed and used to generate OsBZR1 gainof-function plants that can not be created by other available Cas9-engineered base editors. Moreover, we showed that a Cas9-NG-based transcriptional activator efficiently upregulated the expression of endogenous target genes in rice. In addition, we discovered that Cas9-NG recognizes NAC, NTG, NTT, and NCG apart from NG PAM. Together, these findings demonstrate that Cas9-NG can greatly expand the targeting scope of genome-editing tools, showing great potential for targeted genome editing, base editing, and genome regulation in plants.

No abstract

Background Plant genome engineering mediated by various CRISPR-based tools requires specific protospacer adjacent motifs (PAMs), such as the well-performed NGG, NG, and NNG, to initiate target recognition, which notably restricts the editable range of the plant genome. Results In this study, we thoroughly investigate the nuclease activity and the PAM preference of two structurally engineered SpCas9 variants, SpG and SpRY, in transgenic rice. Our study shows that SpG nuclease favors NGD PAMs, albeit less efficiently than the previously described SpCas9-NG, and that SpRY nuclease achieves efficient editing across a wide range of genomic loci, exhibiting a preference of NGD as well as NAN PAMs. Furthermore, SpRY-fused cytidine deaminase hAID*Δ and adenosine deaminase TadA8e are generated, respectively. These constructs efficiently induce C-to-T and A-to-G conversions in the target genes toward various non-canonical PAMs, including non-G PAMs. Remarkably, high-frequency self-editing events (indels and DNA fragments deletion) in the integrated T-DNA fragments as a result of the nuclease activity of SpRY are observed, whereas the self-editing of SpRY nickase-mediated base editor is quite low in transgenic rice lines. Conclusions The broad PAM compatibility of SpRY greatly expands the targeting scope of CRISPR-based tools in plant genome engineering.

No abstract