Improving prime editing with an endogenous small RNA-binding protein

CRISPR-Cas systems have revolutionized genome editing, leading to the development of tools like programmable nucleases, base editors, and prime editors. Prime editors utilize reverse transcription to introduce precise edits with minimal off-target effects, making them highly versatile for genetic modeling, functional genomics, and therapeutic applications. While prime editing offers high precision, its efficiency is often low and variable. To improve prime editing, enhancing its efficiency is a major goal, and a deeper understanding of its interaction with the cellular environment is crucial. Previous studies have focused on factors influencing prime editing efficiency, such as expression levels, stability, localization, and activity of editing components, and the chromatin context of the target locus. However, a comprehensive understanding of the cellular determinants remains elusive. This study aimed to perform genome-scale screens to identify novel cellular factors that impact prime editing efficiency and explore their mechanisms of action.

Several studies have investigated prime editing, particularly focusing on DNA repair processes. However, genome-scale screens—a powerful approach for identifying cellular determinants—have not been extensively used for prime editing or other CRISPR-based technologies. Previous research has shown that small prime edits can be installed more efficiently by suppressing or evading mismatch repair (MMR). This demonstrated how mechanistic understanding could lead to technological improvements. Studies have explored improving prime editing systems, focusing on expression, stability, localization, and activity of editing components, as well as the chromatin context of the targeted loci. The role of MMR in influencing editing outcomes has also been established. This study aimed to broaden the understanding beyond these previously explored aspects.

To identify cellular determinants of prime editing, the researchers developed a scalable prime editing reporter system. This system utilizes a bicistronic mRNA, where the installation of an in-frame ATG start codon by prime editing activates a reporter gene (eGFP or a cell surface protein). Two reporter versions were created, allowing for cell sorting using either fluorescence-activated cell sorting (FACS) or magnetic cell separation (MCS). These reporters were incorporated into K562 CRISPRi cells and used for genome-scale CRISPRi screens with a library targeting 18,905 genes. The screens involved transducing the reporter cells with the CRISPRi library, introducing prime editing components (SaPE2, pegRNA, and nicking sgRNA), and separating reporter-positive and -negative cell populations. Amplicon sequencing was used to determine the relative enrichment or depletion of each sgRNA. In addition to the FACS-based screen, MCS-based screens were also performed using the PE3, PE4, and PE5 prime editing systems. Further validation experiments included using La-targeting CRISPRi sgRNAs, La knockout cells, and La-targeting siRNAs to assess the impact of La depletion on prime editing. To investigate the interaction between La and pegRNAs, synthetic pegRNAs with varying 3' end modifications were designed and tested. Finally, a prime editor protein (PE7) fused to La's N-terminal RNA-binding domain was created and evaluated for its efficiency in enhancing prime editing. Detailed methodology for each step is provided in the supplementary materials.

Genome-scale CRISPRi screens consistently identified the small RNA-binding protein La (encoded by SSB) as the strongest positive regulator of prime editing. Depletion of La significantly impaired prime editing efficiency across different approaches (PE2, PE3, PE4, PE5), edit types (substitutions, insertions, deletions), endogenous loci, and cell types. This effect was independent of MMR. The study found that La interacts functionally with the 3' polyuridylated ends of pegRNAs. Specifically, the N-terminal domain (La(1-194)) of La, which contains the RNA recognition motifs essential for polyU binding, was shown to be sufficient for rescuing prime editing in La knockout cells. Synthetic pegRNAs with 3' polyU tracts that were accessible to La binding showed higher dependence on La for efficient editing, while those with modifications blocking La access were less affected. A prime editor (PE7) with La's N-terminal domain fused to it showed substantial improvement in prime editing efficiency across various loci, cell lines, and edit types. This enhancement was dependent on the RNA-binding activity of the fused La domain. PE7 improved editing efficiencies in U2OS cells by a median of 21.2-fold and 5.5-fold with pegRNAs and epegRNAs, respectively. Importantly, PE7 improved editing efficiency across several disease-relevant targets (HBB, PRNP, PCSK9, IL2RB, CXCR4, CDKL5), showing a median 21.8-fold and 10.8-fold improvement for pegRNAs and epegRNAs in U2OS cells. Finally, PE7 was shown to enhance prime editing in primary human T cells and HSPCs, demonstrating the potential for translation to clinical applications. Small RNA sequencing revealed that La loss destabilizes pegRNAs and epegRNAs, particularly at their 3' ends, suggesting a role for La in pegRNA stability and integrity.

This study significantly advances our understanding of prime editing by identifying La as a key cellular determinant of its efficiency. The findings demonstrate that La's interaction with the 3' polyU tracts of pegRNAs is crucial for its function, likely by enhancing pegRNA stability and protecting them from degradation. The successful development and validation of the PE7 prime editor, incorporating La's RNA-binding domain, provides a substantial improvement in prime editing efficiency. The improved efficiency and versatility of PE7 across various cell types and disease-relevant targets have important implications for therapeutic applications. The study highlights the importance of considering the interaction between editing components and the cellular environment when developing and optimizing gene editing technologies.

This research successfully identified La as a critical cellular factor enhancing prime editing efficiency and developed the PE7 prime editor, which significantly improves editing efficiency by tethering La's RNA-binding domain. PE7 demonstrates enhanced performance in various cell lines and with different edit types, including disease-relevant targets, and even in primary cells. Further optimization of PE7, such as refining linker sequences or exploring in trans La overexpression, holds potential for further improvements in prime editing efficiency. The findings provide crucial insights into prime editing mechanisms and offer promising avenues for future development.

The study primarily focused on specific cell lines (K562, HEK293T, HeLa, U2OS) and a limited set of endogenous loci. While the results were validated in primary cells, further investigation across a broader range of cell types and genomic contexts is needed to assess the generalizability of the findings. The exact mechanism by which La promotes prime editing, beyond its role in pegRNA stability, may warrant further research. The potential off-target effects of PE7, particularly in therapeutic applications, require further investigation. Additionally, long-term effects of PE7 expression and potential interactions with other cellular pathways should be evaluated before clinical translation.