Center for Bioinformatics at KU

As part of our public Seminar Series
http://bioinformatics.ku.edu/seminars

Jan 24, Thu 2007
2:30 pm, 2023 Haworth

John Karanicolas

University of Washington

Design of de Novo Protein-Protein Interfaces

Many proteins carry out multiple complex cellular functions. In order to decipher which portions of the protein surface are responsible for a particular function, it would be desirable to selectively disable a portion of the protein surface. As a first step towards attacking this problem, we have developed a methodology for computationally designing complementary surfaces on a pair of non-interacting proteins to introduce favorable interactions between them.

I will describe two applications of this methodology. The first design uses an amyloid forming hexapeptide from the protein tau as a target, leading to the development of an inhibitor of fibril formation. The second example involves engineering a redesigned ubiquitin molecule by incorporating 14 surface mutations, so that it will bind a redesigned Ankyrin repeat protein harboring 16 surface mutations.

I will outline the computational protocol used to design these interfaces, and will present the experimental characterization of both successful designs. Discussion will focus on our ongoing efforts to design interfaces which bind with higher affinity, as well as extension of this methodology to design proteins that bind endogenous cellular targets.