Computational Design of Novel Proteins for Affinity Reagents Public Deposited

Downloadable Content

Download PDF
Last Modified
  • March 22, 2019
  • Jha, Ramesh K.
    • Affiliation: School of Medicine, Department of Biochemistry and Biophysics
  • Computational protein design is an emerging technology that can pave a way to redesign and create novel proteins which have new functions as affinity reagents and protein therapeutics. Two approaches have been described to create an affinity reagent for activated form of p21-activated kinase 1 (PAK1). A good affinity reagent can be developed into a biosensor for visualizing the activated form of PAK1 inside cells. In the first approach, truncated auto-inhibitory domain (trunc-AID) from PAK1, which is unfolded but binds the activated conformation of PAK1 with an affinity of 4 [mu]M, was redesigned for improvement in structure and function. Using a computational approach implemented within Rosetta molecular modeling program, a 20-residue helix at the N-terminus was created in the N-terminus of trunc-AID that had the PAK1-binding region conserved. The design, AlmostHelix showed gain in structure and solubility but no improvement in binding affinity. Another design, PAcKer, where a 16-residue helix was created at the C-terminus of the trunc-AID and was inserted in a fluorescent protein (CFP) bound the active conformation of PAK1 with an affinity of ~400 nM. CFP-PAcKer did not show any binding to the 'closed' form of PAK1 making it suitable for a biosensor that can detect only the activated form of PAK1 inside cells. A de novo interface design approach was used to create a binding interface on a segment from hyperplastic discs protein ('scaffold') to bind the active conformation of PAK1 ('target'). Using a newly developed protocol DDMI, that is implemented within the Rosetta molecular modeling program and uses rigid-body docking, sequence design, and gradient-based minimization of backbone and side chain torsion angles to design low energy interfaces between the 'scaffold' and 'target', a binder called Spider Roll was designed. Spider Roll bound to PAK1 with a modest affinity of 100 [mu]M. Mutagenesis studies confirmed that the binding interface was consistent with the design model. NMR studies were also consistent with the binding model. Additionally, Spider Roll did not bind the 'closed' PAK1. De novo designed binders that can distinguish between two states of proteins have potential use as biosensors. As a future direction for de novo protein interface design, an attempt was made to create a metal-mediated interaction between a 'scaffold' protein and Ubiquitin as a 'target'. Metal-mediated interactions have advantage over other non-covalent interactions due to strong coordination bond. Experimental testing of multiple designs showed a design (called Spelter) bound with low nanomolar affinity to the metal ion, but any metal-mediated interaction between Spelter and Ubiquitin was not established.
Date of publication
Resource type
Rights statement
  • In Copyright
  • ... in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Biochemistry and Biophysics (Program in Molecular and Cellular Biophysics).
  • Kuhlman, Brian

This work has no parents.