Prime Editing Mechanism

Prime editing represents a remarkable leap forward in genome editing technology. It relies on a prime editing guide RNA (pegRNA) and a prime editing protein comprising a programmable nickase and a reverse transcriptase. The unique combination of these components allows prime editing to insert or delete genetic material with unprecedented accuracy. Unlike traditional CRISPR-Cas9 systems, prime editing does not induce double-strand DNA breaks or require donor DNA templates, reducing the risk of unintended mutations and making it an attractive option for therapeutic applications.

Challenges with Non-M-MLV Editors

While prime editing has shown great promise, one of the early challenges was achieving optimal efficiency, especially when using reverse transcriptases other than the widely used Moloney murine leukemia virus (M-MLV) reverse transcriptase. Prime editors, such as PE2, attempted to enhance editing efficiency by modifying M-MLV reverse transcriptase, but significant limitations persisted.

Selecting Reverse Transcriptases from Various Sources

Recognizing the need for more efficient reverse transcriptases, researchers from the Broad Institute of MIT and Harvard began a comprehensive screening process. They explored 59 reverse transcriptases derived from various evolutionary sources, spanning 14 different categories, as candidate editors for prime editing. This extensive screening revealed 20 reverse transcriptases that exhibited detectable editing activity. Among these, nine were notable for their smaller gene size than M-MLV reverse transcriptase. This finding raised hopes for more compact prime editing tools.

Engineering Reverse Transcriptases

To further optimize reverse transcriptases for prime editing, researchers applied a combination of protein engineering and structural insights. Drawing from prior experience with M-MLV reverse transcriptase, they introduced five key mutations (D200N, T306K, W313F, T330P, and L603W) into the PE2 protein. These mutations were strategically designed to enhance substrate binding and thermal stability, essential factors for efficient editing. The resulting variants were incorporated into various reverse transcriptases, yielding remarkable increases in editing capability.

Notably, the yeast Tf1 reverse transcriptase, which was smaller in size, stood out as a promising candidate. While not surpassing the editing efficiency of PE2, Tf1 approached its levels, making it an attractive option for certain applications. However, challenges remained when dealing with longer and more complex edits.

Increasing Affinity Between Reverse Transcriptases and Substrates

Recognizing the importance of solid binding between reverse transcriptases and their DNA/RNA substrates, researchers delved into structural guidance. They drew inspiration from the Tf1 homolog Ty3 reverse transcriptase and introduced mutations (K118R, S118K, I260L, S297Q, and R288Q) near DNA and RNA substrates in Tf1. These mutations substantially improved editing efficiency, providing further evidence of the crucial role substrate affinity plays in prime editing.

The study also applied similar strategies to the E. coli Ec48 retron reverse transcriptase, which was smaller in size but exhibited lower activity. Using structural predictions generated by AlphaFold2, researchers engineered variants of Ec48 with promising results. The introduction of T189N led to a three-fold improvement in editing efficiency. Further iterations, such as K307R, R378K, L182N, T385R, and R378K, were developed using the same approach, with the rdEc48 variant demonstrating an average prime editing efficiency 8.6 times higher than its wild-type counterpart.

Phage-Assisted Continuous Evolution (PACE) for Prime Editors

Phage-Assisted Continuous Evolution (PACE) technology, initially developed by the David Liu research team in 2011, revolutionized the field of protein-directed evolution. It involves the continuous co-culture of phages and host E. coli, generating mutant libraries through error-prone replication. PACE retains only mutants that align with the desired evolutionary direction, allowing for the creation of tailored proteins with enhanced properties.

Development of PE-PACE Circuit

To harness the power of PACE for prime editing, researchers devised a PE-PACE circuit. This circuit ingeniously linked prime editing activity to phages, enabling the evolution of reverse transcriptase variants optimized for mammalian cell prime editing. By optimizing pegRNA and increasing phage propagation levels, researchers demonstrated the circuit's ability to distinguish reverse transcriptase variants based on prime editing activity.

Evolution of Highly Active AAV Reverse Transcriptase

Building on the advancements in reverse transcriptase evolution, the study combined mutations from Tf1 RT (PE6b) evolved through PE-PACE with those from rdTf1, resulting in the Tf1 variant PE6c. Additionally, researchers pursued parallel evolution of PE2 reverse transcriptase in v1, v2, and v3 circuits, ultimately creating the M-MLV variant with mutations near the polymerase active site, known as PE6d.

Combining PE-PACE and Protein Engineering Techniques

By strategically merging PE-PACE with protein engineering techniques, researchers successfully constructed highly efficient prime editors. Both the Tf1 reverse transcriptase variant (PE6c) and the M-MLV variant (PE6d) matched the size of PEmax△RNaseH while retaining their effectiveness for complex editing tasks.

RTT Secondary Structure Dependency of PE6c, PE6d, and PEmax△RNaseH

To better understand the performance of PE6c, PE6d, and PEmax△RNaseH in structured editing, researchers conducted a comparative analysis. These editors' ability to efficiently handle long edits in a double adeno-associated virus (AAV) delivery system was evaluated. Notably, PE6c and PE6d demonstrated superior performance compared to PEmax△RNaseH in structured editing, underscoring their capability to handle challenging edits.

Importance of RTT Secondary Structure

The study emphasized the significance of RNA template secondary structure (RTT) in determining editing efficiency. When RTTs contained hairpin structures, PE6d significantly outperformed PEmax△RNaseH in editing efficiency. In the absence of hairpin structures, both editors performed similarly. This observation highlighted the importance of understanding RTT secondary structure for optimizing editor performance.

Performance in Dual Guide Editing

The researchers further evaluated the efficacy of PE6c and PE6d in dual guide editing, demonstrating their ability to outperform PEmax△RNaseH. In specific tests at the mouse Rosa26 locus, PEmax△RNaseH generated a substantial amount of insertions and deletions (indels), while PE6c and PE6d increased editing efficiency while reducing indels. These results underscored the potential of PE6c and PE6d for complex editing tasks with minimal off-target effects.

Predictive Method Based on RTT and PBS Folding Free Energy

The researchers also established a predictive method based on the folding free energy of the RTT and primer binding site (PBS) to determine when using PE6d instead of PEmax△RNaseH could yield better editing efficiency. Stronger folding free energy was associated with greater efficiency when using PE6d, offering a valuable predictive tool for optimizing editing protocols.

Summary

In conclusion, the recent advancements in prime editing technology, particularly the development of compact, efficient prime editors through Phage-Assisted Continuous Evolution (PACE), represent a remarkable stride in the field of genome modification. These innovations offer new possibilities for precise genetic corrections, potentially ushering in groundbreaking developments in treating genetic diseases and advancing genetic research.

The collaboration between structural biology, protein engineering, and evolutionary techniques continues to push the boundaries of what is achievable in the field of genome editing. The combined efforts of researchers have yielded a powerful toolset for precise genome editing with applications across various biological contexts, including gene therapy and functional genomics research. These findings hold great promise for the future of genetic engineering and its potential to address previously insurmountable challenges in the realm of life sciences.

Reference

[1] Jordan L. Doman, et al. Phage-assisted evolution and protein engineering yield compact, efficient prime editors. Cell 186, 3983–4002 (2023)

[2] Blum, T. R., et al. Phage-Assisted Evolution of Selective Botulinum Neurotoxin Proteases with Reprogrammed Substrate Specificity. Science 371, 803-810 (2021)

[3] Chen. P. J., et al. Enhanced Prime Editing Systems by Manipulating Cellular Determinants of Editing Outcomes. Cell 22, 5635–5652 (2021)