Efficient plant genome engineering using a probiotic sourced CRISPR-Cas9 system

CRISPR-Cas systems have revolutionized genome editing, with *Streptococcus pyogenes* Cas9 (SpCas9) being the most widely used. However, SpCas9's 5'-NGG-3' PAM requirement limits its targeting scope. Efforts to overcome this limitation have focused on discovering Cas9 orthologs with alternative PAM specificities or engineering SpCas9 variants with relaxed PAM requirements. While some progress has been made, these alternatives often exhibit compromised editing efficiency. The CRISPR-Cas12a system offers an alternative with a 5'-TTTV-3' PAM, but it lacks the versatility of CRISPR-Cas9, particularly in the development of sophisticated tools like base editors and prime editors. This study aimed to identify and characterize a novel CRISPR-Cas9 system from a probiotic source to provide a safer and more efficient alternative for plant genome engineering. The use of a probiotic source also addresses concerns about using human pathogens in genome editing, which could be beneficial for public acceptance of genome-edited crops.

The literature extensively documents the use of SpCas9 and its variants for plant genome editing, including base editing, prime editing, and gene regulation. Various Cas9 orthologs with alternative PAM specificities have been explored, but many exhibit limited editing activity or complex PAM sequences. Similarly, engineered SpCas9 variants with relaxed PAMs have shown improved targeting scope but often with reduced efficiency. The CRISPR-Cas12a system, while useful for multiplexed editing and promoter editing, lacks the versatility of Cas9 systems. The development of Cas12a-based base editors has been recent and less developed than Cas9-based counterparts. This research gap motivated the search for a novel, efficient, and safe Cas9 enzyme from a probiotic source.

The researchers employed a three-step in silico approach for identifying potential CRISPR-Cas9 systems: Cas protein identification, CRISPR array identification, and PAM prediction. A total of 33,825 proteomes were analyzed, leading to the identification of 42 CRISPR-Cas9 candidates with diverse PAM predictions. The researchers selected four candidates (including LrCas9 from *Lactobacillus rhamnosus*) for experimental validation in rice protoplasts. LrCas9 exhibited high editing efficiency, prompting further characterization. The PAM preference of LrCas9 (5'-NGAAA-3') was confirmed using a bacterial PAM depletion assay. The study optimized sgRNA scaffolds for LrCas9 and tested its efficiency at various spacer lengths and temperatures. Comparisons were made with LbCas12a, SpCas9-NG, and SpRY at identical target sequences in rice, wheat, tomato, and *Larix* protoplasts. The development of LrCas9-based CBE and ABE involved fusing LrCas9 nickase with appropriate deaminases. CRISPRi and CRISPRa systems were engineered by fusing dLrCas9 with repressor and activator domains, respectively. Off-target analysis was conducted using amplicon deep sequencing, GUIDE-seq, and direct genotyping of stable rice lines. Multiplexed genome editing in stable rice plants was demonstrated using tRNA-based sgRNA processing. Promoter editing was performed targeting the *OsWx* promoter, and the resultant amylose content was assessed.

LrCas9, sourced from the probiotic *Lactobacillus rhamnosus*, demonstrated superior editing efficiency compared to LbCas12a, SpCas9-NG, and SpRY across various plant species (rice, wheat, tomato, and *Larix*) when targeting identical sequences. LrCas9 showed a strong preference for the 5'-NGAAA-3' PAM sequence. The study successfully developed functional LrCas9-derived cytosine base editors (CBEs) and adenine base editors (ABEs) in rice and wheat. Multiplexed gene editing with LrCas9 was highly efficient, leading to simultaneous editing of multiple target genes and the generation of large chromosomal deletions. Targeted promoter deletion using LrCas9 resulted in significant reduction of *OsWx* expression, leading to a substantial decrease in amylose content in rice seeds without significantly affecting total starch content. Comprehensive off-target analysis using three independent methods (amplicon deep sequencing with mismatched spacers, GUIDE-seq, and direct genotyping of stable lines) demonstrated high specificity of LrCas9. Efficient CRISPRi and CRISPRa systems based on LrCas9 were developed, achieving potent transcriptional repression and activation, respectively. Notably, the nuclease-active LrCas9-TV system demonstrated significantly higher activation potency compared to the dLrCas9-TV system.

The findings demonstrate that LrCas9 is a highly efficient and specific genome editing tool suitable for diverse applications in plants. Its superior performance compared to existing systems, combined with its source from a probiotic organism, addresses both efficacy and safety concerns in plant genome engineering. The development of LrCas9-based base editors and gene regulation systems expands the capabilities of this platform. The high efficiency of multiplexed editing and promoter editing makes LrCas9 particularly valuable for engineering quantitative traits. The high specificity demonstrated by LrCas9 minimizes the risk of unintended off-target effects. The use of a probiotic source also addresses public concerns about the use of human pathogens in genome editing, potentially paving the way for wider acceptance of genome-edited crops.

This study establishes LrCas9 as a powerful and versatile genome-engineering tool for plant applications. Its superior performance, safety, and expanded capabilities make it a valuable addition to the existing CRISPR toolkit. Future research could explore further optimization of LrCas9 through protein engineering and the development of LrCas9-based prime editing systems for even greater precision in plant genome modification.

While the study demonstrates high specificity, the off-target analysis was primarily focused on rice. Further research is needed to confirm the specificity of LrCas9 in other plant species. The study primarily used protoplasts for initial testing; while stable transgenic plants were also used, the long-term effects of LrCas9-mediated genome editing need further investigation. Although a highly efficient CRISPRa system was developed, the mechanism warrants further investigation. Finally, only a limited number of target sites were used in each species. Future research may require broader testing of different target sites in more plant species.