The Degreaser: spot cleaning sequences to maximize protein secretion

Abstract

As computational protein design enables the creation of novel features, forms and functions, such as designed protein nanoparticles, expression and secretion from eukaryotic cells becomes advantageous for the generation of new biologics. However, many designed proteins secrete poorly. Because hydrophobic interfaces are designed into otherwise soluble proteins, many of these proteins gain hydrophobic segments that may be interpreted by cellular membrane insertion machinery to be transmembrane domains. To address this, we develop a computational method based on a transmembrane insertion prediction model and Rosetta: the Degreaser. We use the Degreaser to identify cryptic transmembrane domains and design them away without compromising the originally-designed protein. Retroactive application of the Degreaser to previously designed nanoparticle components and nanoparticles considerably improves secretion. Modular integration of the Degreaser into design pipelines and incorporation of the Degreaser in large-scale protein design results in proteins that secrete robustly. These secretion-optimized proteins represent the first generation of proteins that are specifically designed for maximal secretion. Future generations of designed proteins can incorporate other secretion optimization features, especially if they can be selected from high-throughput, multiplex screens.