Viral infection results in thousands of deaths and an enormous humanitarian burden every year, yet
unlike antibiotics for bacterial infection, very few antivirals are available. At present, few methods exist for
generating effective antivirals. The increasing availability of atomic resolution structural information of
various viral surface proteins promises to chance this. The overall objective of this application is to use the
surface glycoproteins of Paramyxoviruses (PMV) as a model system to generate design methodologies
that will take advantage of structural features present in a broad range of viruses, resulting in a robust
platform for the design of new therapeutics, diagnostics and immunogens for vaccination. PMVs are an
ideal model system as their family members have the same fold for receptor recognition yet bind to very
different host cell receptors.
We recently demonstrated that computational protein design can be used to generate de novo antivirals
that broadly neutralize diverse strains of influenza. These computer-generated proteins can also function
as highly sensitive diagnostics. Guided by these results, the following specific aims will be pursued: (i)
Develop general design strategies to target virus:host cell receptor interactions and design antivirals using
Hendra and Nipah Viruses as model systems; (ii) inhibit membrane fusion of RSV by targeting the
intermediate fusion states; and (iii) selectively stabilize the pre- and post-fusion state stabilization of the F-
protein of RSV and probe their contributions to infectivity and vaccine design.
The first aim is based on the observation that many receptor-binding sites of enveloped viruses lay
within a recessed pocket, enabling evasion from the immune system. Computational design strategies
which specifically target pockets will enable the development a robust algorithm to generate antiviral
proteins which bind at these sites. The second aim is based on the hypothesis that the post-fusion
structure of viral surface proteins provides the blueprint to targeting their transition state. Small proteins will
be designed to molecularly “jam” the 3-helical core structure that is common to most type I fusion proteins
and therefore will be provide a general method to inhibit type I fusion proteins, which include viruses such
as HIV-1, Ebola, SARS and others. Lastly, the objective of aim three is to simultaneously model the pre-
and post-fusion states of the F-protein of RSV to generate variants to favor one state over the other.
Variants will be assayed for changes in infectivity. The trapped pre-fusion state stabilized by disfavoring the
post-fusion state will provide the basis for a new angle on immunogen design. If successful, data on
designs will be fed back into the developed algorithm, leading to rapid development of new antivirals
against emerging epidemics.