Access denied: researchers identify a new strategy to block viral infection

Posted: December 4th, 2014

Many historically devastating viral pathogens (including influenza, HIV, and Ebola) lurk behind a superficially innocuous lipid membrane, typically acquired from the host organism, that may reduce immune system attention by mimicking normal cell surfaces.    These so-called enveloped viruses rely on their own glycoproteins to effect fusion between the viral envelope and a target cell membrane in order to begin a new infection cycle. The fusion proteins function in part by binding a relatively short ligand peptide to the surface of a trimeric coiled coil subdomain, and inhibitors that block this ligand binding site are known to ameliorate or block infection.

The native ligand sequences are good starting points to discover new antivirals, and indeed a segment of that sequence from the HIV fusion protein gp41 is now an FDA-approved anti-AIDS drug (marketed as Fuzeon). But short peptide sequences typically make lousy drugs: they are very expensive to make, and the body’s natural defense systems tend to chew them up rather quickly, limiting ‘bioavailability’ (the amount of drug per dose available to do biological work). There thus exists a demand for a robust protein scaffold, to which these sequences can easily be grafted in such a way as to present a viable binding partner. Such larger proteins are easily prepared by modern molecular biology methods, and tend to exhibit more prolonged stability in environments like human serum.

This approach has recently been validated by graduate students Suzanne Walker, Rachel Tennyson, and Alex Chapman, working under the direction of Chemistry professors Alan Kennan and Brian McNaughton. The work was recently published in ChemBioChem ( In it, the authors demonstrated that a strategy of ‘helix grafting’ allowed attachment of a helical gp41 epitope onto the native helix of a protein called a GLUE domain. This hybrid GLUE-gp41 construct potently and selectively binds to a standard model of the coiled-coil receptor, with efficacy comparable to (or even slightly better than) the native sequence. As expected, the hybrid protein was easily expressed in bacterial E. coli cells, and exhibits a serum half-life well in excess of that observed for Fuzeon. Thus even this proof-of-concept non-optimized sequence has the potential to significantly disrupt HIV infection. Current work focuses on optimization of the binding interface, as well as extending the grafted ligand display concept to other viral systems like Ebola.