Related Experiment Video
Updated: May 23, 2026
09:16
In Vitro Directed Evolution of a Restriction Endonuclease with More Stringent Specificity
Published on: March 25, 2020
AI-guided redesign of laboratory-evolved reverse transcriptases enhances prime editing.
Y Allen Tao1,2,3, Holt A Sakai1,2,3, Allen Y Jiang1,2,3
1Merkin Institute of Transformative Technologies in Healthcare, The Broad Institute of MIT and Harvard, Cambridge, MA, USA.
Nature Biotechnology
|May 21, 2026
Summary
Protein engineering for prime editors improved efficiency but reduced stability. Artificial intelligence redesigned reverse transcriptase domains, enhancing stability and boosting prime editor performance in cells and mice.
Area of Science:
- Biochemistry
- Molecular Biology
- Bioengineering
Background:
- Protein engineering and laboratory evolution optimize prime editors.
- Previous optimization efforts improved prime editor efficiency but compromised protein stability and expression levels, limiting overall performance.
Purpose of the Study:
- To address limitations in prime editor stability and expression.
- To enhance prime editor performance using structure-informed AI-guided methods.
Main Methods:
- Applied artificial intelligence (AI)-guided methods, including the ProteinMPNN inverse-folding network, to redesign reverse transcriptase (RT) domains of prime editors.
- Preserved essential catalytic regions while introducing 30-163 amino acid substitutions in redesigned RTs.
Main Results:
- Redesigned RTs exhibited enhanced folding stability and soluble expression.
- Achieved up to twofold higher intracellular prime editor protein levels after mRNA delivery.
- Redesigned PE8 prime editors showed enhanced editing efficiencies in human primary cells and various delivery modalities.
- Demonstrated up to 2.9-fold higher editing efficiency in mice compared to PE6, PE7, and PEmax prime editors.
Conclusions:
- AI-guided redesign of RT domains significantly improves prime editor stability and expression.
- The developed approach enhances prime editor performance in diverse cellular contexts and in vivo.
- This strategy offers a generalizable method to augment laboratory evolution for improved genome editing agents.
