MutDock: A computational docking approach for fixed-backbone protein scaffold design.

Area of Science:

• Protein engineering and computational biology
• Biomolecular design and drug discovery
• Structural bioinformatics

Background:

• Therapeutic antibodies face production and design challenges, leading to the development of smaller alternative binding scaffolds.
• Alternative scaffolds recognize targets via surface mutations without altering backbone structure, but computational design lags behind antibodies.
• Existing computational methods like dock-and-mutate have limitations due to fixed scaffold representations.

Purpose of the Study:

• To develop a novel computational approach, MutDock, for simultaneous docking and mutation of fixed-backbone scaffolds.
• To enhance the identification of superior binding poses by considering scaffold mutations during docking.
• To improve the discovery of novel binders based on smaller protein scaffolds.

Main Methods:

• MutDock employs a two-step process: pairwise distance alignment of hydrogen bonds and mutation of unfavorable residues.
• The method considers native and mutated rotamers of scaffold residues and compatible epitope atoms.
• MutDock was applied to Affibodies and DARPins, docking them with ten random antigens.

Main Results:

• MutDock identified docked poses with higher or comparable binding energies compared to ZDOCK and HADDOCK.
• The approach successfully docked smaller, fixed-backbone protein scaffolds with antigens.
• Binding energies of MutDock poses were minimized and compared against established docking software.

Conclusions:

• MutDock offers a novel computational strategy for designing protein binders with fixed-backbone scaffolds.
• The method advances the computational design of alternative binding proteins beyond traditional antibody design.
• This work facilitates the discovery of new therapeutic binders using smaller, engineered protein scaffolds.