Jan 24, 2024

In the era of CRISPR/Cas9 technology, gene editing techniques have evolved rapidly, with base editing technology emerging as a mature and targeted approach. While most base editing breakthroughs have concentrated on deaminases, recent strides have been made by a research team from the Tianjin Institute of Industrial Biotechnology, Chinese Academy of Sciences. Their innovative work introduces a non-deaminase base editor based on glycosyltransferases.

Senior author of the study: Dr. Changhao Bi

A Novel Approach to Base Editing

Rather than targeting the amino group of the base through deaminases, the research team is taking a different approach, aiming to achieve editing by targeting the ribose of the nucleotide's five-carbon sugar core. The uracil-DNA glycosylase (UNG) present in cells is an excellent starting point for research, as it can hydrolyze the glycosidic bond between uracil and the sugar-phosphate backbone. If suitable cytosine DNA glycosylase (CDG) and thymine DNA glycosylase (TDG) can be found and combined with the Cas9 endonuclease, it may be possible to create a simple base editing system.

Contrary to initial optimistic expectations, preliminary experiments reveal the incapacity of these two systems to detect the manifestation of base editing in Escherichia coli. The study posits this limitation as a likely consequence of the diminished activity of CDG and TDG. Consequently, a targeted initiative was undertaken to enhance the functional efficacy of the initial CDG and TDG versions through a process of directed evolution.

Following 33 iterative rounds of directed evolution within the Escherichia coli system, the research team successfully discerns the emergence of the two most proficient mutants, denoted as CDG4-Cas9 and TDG3-Cas9. Notably, these mutants exhibit a noteworthy capability for effecting C/T editing within Escherichia coli.

Subsequently, the research team introduced these two systems into human cells and successfully observed various types of C/T base editing. They also identified the corresponding preferred target sequences. In HEK293 cells, the base substitutions induced by DAF-CBE were mostly C to G and C to T, while those induced by DAF-TBE were mostly T to G and T to C.

Base Editors with Different Strengths

After demonstrating the feasibility of DAF-BE, researchers compared it with other base editing tools. Previously, CGBE (C to G base editor) and PE (Prime Editor) were both reported to achieve specific edits, with the latter theoretically capable of editing all four bases.

Compared to four other CGBEs, DAF-CBE showed significantly higher editing efficiency at specific sites, while the other four CGBEs had better editing efficiency at other sites. PE and DAF-TBE also demonstrated strengths in T to G editing, making it challenging to determine superiority.

Regarding off-target effects, DAF-CBE exhibited a noticeable advantage compared to the other two CBEs. Meanwhile, DAF-TBE did not show a significant increase in off-target effects compared to the control group.

Subsequently, researchers validated the clinical application potential of DAF-BE. Using DAF-BE, they successfully introduced pathogenic single-base mutations at specified sites in human-induced pluripotent stem cells (hiPSCs). After replacing the original Cas9 with the improved SpRY endonuclease, DAF-BE efficiently corrected pathological mutations. The correction efficiency of SpRY-DAF-CBE for specific disease mutations reached up to 36.4%.

These data collectively indicate that DAF-BEs can further enrich our toolbox for base editing and possess significant clinical application value.

Further Modification

After validating the concept and potential of DAF-BEs, researchers further modified and upgraded DAF-BEs to achieve higher stability and editing efficiency. By optimizing the connection between glycosyltransferases and Cas9 and introducing circular permutations into Cas9, the researchers elevated DAF-BEs to a new level.

In testing the ten CDG4 variants, different variants exhibited differentiated editing purity and editing windows. Among them, CDG4-CP1249 showed an increase in editing purity from 38.7% to 58.4% at one primary editing site, and the C to G editing purity at another primary editing site jumped from 2.9% to 66.6%. After editing 33 sites using CDG4-CP1249, researchers found a significant improvement in average editing efficiency and purity. This superior version was named DAF-CBE2.

A similar phenomenon was reproduced in DAF-TBE. Although the improvement in editing efficiency was not as remarkable as DAF-CBE, through "magic modification" of the first-generation DAF-TBE, researchers nearly doubled the T to G editing purity, resulting in the iterated DAF-TBE2.

Discussion

The emergence of DAF-CBE2 and DAF-TBE2 suggests that DAF-BEs have great potential for improvement and endless possibilities. At the same time, compared to other BEs, DAF-BEs consist only of CDG/TDA and Cas9, with a smaller molecular weight, providing greater flexibility in usage. Their lower off-target effects and therapeutic potential in human cells will likely earn them a larger stage.

We hold a conviction that the integration of DAF-BEs into our repertoire will profoundly enhance our base editing capabilities, expediting the widespread application of base editing methodologies across diverse industries.

Reference

[1] Ye, L., Zhao, D., Li, J. et al. Glycosylase-based base editors for efficient T-to-G and C-to-G editing in mammalian cells. Nat Biotechnol (2024). https://doi.org/10.1038/s41587-023-02050-w.