Book of Abstracts: Albany 2009

category image Albany 2009
Conversation 16
June 16-20 2009
© Adenine Press (2008)

Computer-based Design of Protein-Protein Interactions

Strategies have been developed for three problems in protein interface design: 1) increasing the affinity of naturally occurring interactions, 2) redesigning protein-protein binding specificities, and 3) designing interactions from scratch. All three approaches make use of the sequence and backbone optimization protocols in the molecular modeling program Rosetta. In general, designs that make use of hydrophobic interactions have been more successful than designs that rely on novel hydrogen bonding networks. This is not ideal as incorporating hydrophobic residues on to the surface of proteins can result in non-specific binding and aggregation. To design more polar interfaces we have developed a protocol that combines molecular modeling with combinatorial screening. Independent sequence optimization trajectories are performed on a large set of perturbed interfaces, and then used to generate amino acid profiles for each residue at the interface. Libraries based on these profiles are experimentally screened for binding. This strategy has been used in one case to design a hydrogen bonding network around a novel histidine residue placed at the center of an interface. The designed interaction has an equilibrium dissociation constant of 30 nM.

Gurkan Guntas
Carrie Purbeck
Deanne Sammond
Ziad Eletr
Ramesh Jha
Brian Kuhlman*

Department of Biochemistry and Biophysics
University of North Carolina
Chapel Hill, NC, 27599-7260

ph: 919-843-0188
fx: 919-966-2852