Cell Research · 2024 · 29 citations · 63 references
Life SciencesBiochemistryMolecular BiologyDrug DiscoveryProtein-protein InteractionsStructural BiologyLigand BindingProtein DesignProtein StructureProtein-ligand BindingProtein-protein RecognitionReceptor-ligand InteractionsD-amino Acid PeptidesDe Novo Protein Binder DesignProtein-protein Interaction DesignProtein-protein Recognition MotifsStructure-based Binding MechanismsStructure-based De Novo DesignTherapeutic PeptidesTropomyosin Receptor Kinases
Abiotic D-proteins that selectively bind to natural L-proteins have gained significant biotechnological interest. However, the underlying structural principles governing such heterochiral protein-protein interactions remain largely unknown. In this study, we present the de novo design of D-proteins consisting of 50-65 residues, aiming to target specific surface regions of L-proteins or L-peptides. Our designer D-protein binders exhibit nanomolar affinity toward an artificial L-peptide, as well as two naturally occurring proteins of therapeutic significance: the D5 domain of human tropomyosin receptor kinase A (TrkA) and human interleukin-6 (IL-6). Notably, these D-protein binders demonstrate high enantiomeric specificity and target specificity. In cell-based experiments, designer D-protein binders effectively inhibited the downstream signaling of TrkA and IL-6 with high potency. Moreover, these binders exhibited remarkable thermal stability and resistance to protease degradation. Crystal structure of the designed heterochiral D-protein-L-peptide complex, obtained at a resolution of 2.0 Å, closely resembled the design model, indicating that the computational method employed is highly accurate. Furthermore, the crystal structure provides valuable information regarding the interactions between helical L-peptides and D-proteins, particularly elucidating a novel mode of heterochiral helix-helix interactions. Leveraging the design of D-proteins specifically targeting L-peptides or L-proteins opens up avenues for systematic exploration of the mirror-image protein universe, paving the way for a diverse range of applications.
63
Highly accurate protein structure prediction with AlphaFold
John Jumper, Alexander Pritzel, Tim Green et al. · Nature · 2021 · 45.2K citations · Full text
PatchDock and SymmDock: servers for rigid and symmetric docking
Dina Schneidman‐Duhovny, Yoel Inbar, Ruth Nussinov et al. · Nucleic Acids Research · 2005 · 3.1K citations · Full text