(331a) Computational Design of Proteins Targeting the Conserved Stem Region of Influenza Hemagglutinin

Whether our knowledge of the various forces that hold dissociable protein complexes together is sufficient to design de novo proteins that can bind an arbitrary target is a grand challenge in molecular biology.  Success would show that we could traverse the gulf that separates the conception of the function of a protein to its primary sequence.  Additionally, success would show a conceivable path for the computational-based development of antibodies to bind to nearly any target.  In this talk, I will present computational methods that were used to design two proteins that bind a neutralizing epitope of the influenza hemagglutinin (HA) from the 1918 H1N1 pandemic virus (Science, in press).  After affinity maturation, two of the designed proteins bind H1 and H5 subtype HAs with low-nanomolar affinity. The crystal structure of one design in complex with 1918/H1 HA revealed that the actual binding interface is nearly identical to that in the computational design model.  Such designed proteins may be useful for both diagnostics and therapeutics.