Search

If we truly understand proteins, we should be able to design functional proteins purposefully from scratch. While the de novo design of proteins has seen many successes1–11, no small molecule ligand- or organic cofactor-binding protein has been designed entirely from first principles to achieve i) a unique structure and ii) a predetermined binding-site geometry with sub-Å accuracy. Such achievements are prerequisites for the design of proteins that control and enable complex reaction trajectories, where the relative placements of cofactors, substrates, and protein side chains must be established within the length scale of a chemical bond. Here, we develop and test a strategy for design of small molecule-binding proteins, based on the concept that the entire protein contributes to establishing the binding geometry of a ligand12–15. Hence, what are traditionally considered as separate sectors – the hydrophobic core and ligand-binding site – we treat as an inseparable unit. We utilize flexible backbone sequence design of a parametrically defined protein template to simultaneously pack the protein interior both proximal to and remote from the ligand-binding site. Thus, tight interdigitation of core side chains quite removed from the binding site structurally restrains the first- and second-shell packing around the ligand. We apply this principle to the decades-old problem of structural non-uniqueness in de novo-designed heme-binding proteins16. We designed a novel protein, PS1, which binds a highly electron-deficient, non-natural porphyrin at temperatures up to 100 °C. The high-resolution structure of holo-PS1 is in sub-Å agreement with the design. The structure of apo-PS1 retains the remote core packing of the holo, predisposing a flexible binding region for the desired ligand-binding geometry. Our results reveal the unification of core packing and binding site definition as an essential principle of ligand-binding protein design.