Announcing the winners of the Winter Protein Design Games 2024: Unleashing Innovation in Enzyme Engineering

The Winter RosettaCon organizers want to thank Liberum Bio for sponsoring the experimental validation of the designs that they carried out in 8 days starting from digital sequences! Liberum Bio produces proteins at high yields in cell-free systems at a fraction of the usual cost and time via highly engineered strains and state-of-the art reaction mixtures.

At Winter RosettaCon 2024 in Boston, the Rosetta Commons in partnership with Liberum Biotech hosted the inaugural Winter Protein Design Games, focusing on the design of Tobacco Etch Virus (TEV) protease variants. TEV protease is commonly used as a tool for sequence-specific protein cleavage, facilitating the purification of proteins without affinity tags. The competition challenged cross-institutional participating teams to engineer TEV proteases that exhibited enhanced substrate cleavage efficiency over the wild-type enzyme. 

The competition was straightforward: design TEV protease variants that outperform the wild type in cleaving a specific peptide substrate. Participants were encouraged to innovate by creating de novo designs, generating novel mutations to the target, or curating known mutations from the literature. In less than ten days, Liberum Bio then synthesized and amplified the designs, and expressed them using a cell-free protein synthesis system. Enzymes were evaluated based on their catalytic efficiency using a fluorophore-quencher labeled substrate. Designed sequences are publicly available. 

Seven cross-institutional teams signed up for the competition at Winter RosettaCon with five teams submitting final designs. The winning team will share a cash prize of $5,000 and will present their results at Summer RosettaCon 2024. 

The teams utilized a range of computational approaches such as ThermoMPNN, PROSS, RFDiffusion, AlphaFold, homology modeling, molecular dynamics simulations, and combined them with known mutations from the literature known to improve solubility and catalytic activity. 

Team 6 won the competition by designing the TEV protease variants with the greatest activity, achieving significant improvements in enzyme efficiency in a head-to-head comparison against wildtype TEV protease. Three of their four designs had the highest activity of all submitted designs, all three of which had higher activity than wildtype, and two of them had even higher activity than the positive control with already-enhanced activity. 

Enzymatic activity of the submitted designs, experimentally synthesized, expressed, and tested by Liberum Bio

The winning strategy from Team 6 involved integrating information from various sources. An initial literature review provided mutations that enhanced binding, catalytic activity, and solubility. Using crystal structures, AlphaFold2 predictions, and homology modeling, the team then modeled the full-length TEV protease bound to the target peptide (ENLYFQG). Following, the team redesigned the active site sequence with ProteinMPNN and conducted molecular dynamics (MD) simulations in both the apo and peptide-bound states to identify key gatekeeper residues. This led to several designs:

• Design 1: Combined established catalytic and solubility mutations from the literature.

• Design 2: Included sets of catalytic and solubility mutations, enhanced by the sequence redesign.

• Design 3: Introduced mutations at MD-identified gatekeeper residues to improve enzyme access and efficiency.

• Design 4: Truncated the C-terminus to enhance the diffusion of the target peptide into the active site at the cost of specificity.

While Team 6's success underscores the value of their methodology, the contributions from other teams remain significant. These methodologies not only broadened the scope of the competition but also enriched the collective understanding of protein design possibilities. The diverse approaches taken by the teams emphasize the competition’s role in fostering innovation and collaboration in the scientific community.

The TEV Protease Winter Protein Design Games marks a collaborative advancement in protein engineering, demonstrating the powerful synergy between various computational design and modeling tools, and state-of-the-art biochemical methods. The competition not only provided a platform for showcasing and testing innovative ideas but also set the stage for future challenges that will continue to push the boundaries of what can be achieved in enzyme design.

The organizers thank all participating teams and want to send a huge "Congratulations!!!" to the winning team, Team 6. Participants from Team 6 are: 

Russell Ault (Children's Hospital of Philadelphia)

Sam Garfinkle (University of Pennsylvania)

Andrew Nelson (University of Pennsylvania)

Colby Agostino (The Wistar Institute)

Shahlo Solieva (The Wistar Institute)

Michaela Helble (University of Pennsylvania)

Caroline Davis (University of Colorado)

Note: None of the conference organizers participated in team organization, computational design of variants, experimental validation or judging of the results.