Rational design of synthetic proteins using a genome-scale CRISPR screen.

Area of Science:

• Synthetic biology
• Genome engineering
• Protein engineering

Background:

• Deep learning has advanced protein structure prediction, but understanding protein function is crucial for designing novel proteins for complex biological tasks.
• Current gene editing technologies face limitations in efficiency and targeting specific genomic loci.

Purpose of the Study:

• To adopt a function-first approach for the rational design of synthetic proteins.
• To identify native human proteins that can enhance precise gene editing.
• To develop novel synthetic genome editors (TruEditors) for improved gene editing efficiency and versatility.

Main Methods:

• Genome-scale CRISPR activation was used to overexpress approximately 19,000 human proteins and measure their impact on homology-directed repair (HDR).
• Top-performing native proteins were fused with the Cas9 nuclease to create synthetic genome editors (TruEditors).
• Affinity proteomics was employed to investigate the mechanism of action of the synthetic proteins.

Main Results:

• Over 800 native human proteins were identified that promote homology-directed repair.
• Twelve unique TruEditors were developed, demonstrating improved precise gene editing across diverse cell types and genomic loci.
• TruEditors delivered via mRNA significantly enhanced chimeric antigen receptor (CAR) insertion in human T cells and improved precise editing in human pluripotent stem cells.

Conclusions:

• Genome-wide protein overexpression screens can effectively guide the rational design of synthetic proteins for specific biological functions.
• The developed TruEditors represent a significant advancement in precise gene editing technology.
• This work opens new avenues for engineering proteins to perform complex biological tasks, with broad implications for gene therapy and regenerative medicine.