Systematic Conformation-to-Phenotype Mapping via Limited Protein Sequencing.

Non-native conformations drive protein misfolding diseases, frustrate bioengineering efforts, and fuel molecular evolution. No current experimental technique is well-suited for elucidating them, and computational techniques lack training data. Deep mutational scanning can comprehensively map complex protein fitness landscapes but not their conformational landscapes. We describe an approach to systematically discover, stabilize, and resolve native and non-native conformations of a protein, in vitroor ex vivo, and directly link them to their molecular, organismal, or evolutionary phenotypes, using high-throughput disulfide scanning (HTDS) of the entire protein. To reveal characteristic disulfide bonding patterns of chromatographically resolved conformers, we devised a limited protein deep-sequencing method that identifies the exact sequence positions of two Cys residues simultaneously within the same polypeptide chain. HTDS of the E. coli periplasmic chaperone HdeA revealed phenotypically distinct classes of disordered hydrophobic conformers. HTDS can determine phenotypic effects of non-native conformations for many proteins that function in disulfide-permissive environments.