Colorado State University

The picornaviruses are a family of small positive sense single stranded RNA viruses that cause a wide range of diseases at an annual cost well into the hundreds of million dollars. Members include the paralyzing poliovirus, the heart disease causing coxsackie B3 virus, coxsackie A viruses that cause hand-foot-and-mouth disease in children, foot and mouth disease virus that is a major livestock pathogen, and the rhinoviruses that cause more than half the occurrences of the common cold. These viruses share a common life cycle where their RNA replication and viral assembly occurs in large membrane anchored replication centers assembled on the surfaces of vesicles derived from the endoplasmic reticulum and increased phospholipid synthesis as a result of viral infection.

There is mounting evidence in several picornaviruses that the polymerase and/or its immediate precursors are responsible for the assembly of these replication centers. Poliovirus polymerase has been shown by to assemble into large sheet structures along a protein-protein interface that was initially identified in a partial crystal structure of 3Dpol. Mutations that disrupt this interface also disrupt viral RNA binding and synthesis in vitro and affect viral viability in vivo, indicating that this interaction is important for proper viral replication.

The viral replication process is driven by a virally encoded RNA dependent RNA polymerase, the 3Dpol protein, that is responsible the synthesis of both (-) and (+) strand viral RNA. Like all picornaviral proteins, the polymerase is generated by proteolytic cleavage of the large ≈250 KDa viral polyprotein that is encoded by the single open reading frame in the viral genome. In the case of 3Dpol, polyprotein processing leads to an activation of the polymerase activity because the 3CDpro precursor protein is only active as a protease. 3CDpro has been implicated as a control molecule responsible for the proper initiation of both negative and positive strand viral RNA synthesis as it forms multiple complexes with RNA secondary structures and 3Dpol at different stages of the viral replication process.

Our contributions to understanding the molecular biology of picornaviral replication are largely focused on structure-function studies of using purified viral proteins and RNA substrates, and then collaborating with virologists to establish how polymerase structure affects viral replication in tissue culture and animal systems. We began our studies by solving the crystal structures of 3Dpol from poliovirus (2004) and coxsackievirus (2008) that showed how proteolytic processing leads to polymerase activation. We then solved the structures of poliovirus, coxsackievirus, and rhinovirus 3Dpol-RNA elongation complexes (ECs), which showed that the (+)-strand RNA virus polymerases use a novel palm domain based conformational change to close their active sites for catalysis (2010, 2013). This mode of active site closure is distinctly different from that found in other classes of replicative polymerases which utilize fingers domain movements to position the NTP for catalysis, and we hypothesized that the palm based movement is likely the evolutionary origin of the low fidelity and "quasi-species" type swarms of variants that are a hallmark of (+) strand RNA viruses.

To more explicitly test this, we used protein engineering to design a set of polymerase mutations predicted to alter replication fidelity and showed, in collaboration with Marco Vignuzzi's group, that the resulting low-fidelity variant viruses had almost 3-fold higher mutation rates and were attenuated for growth in vivo (2012). This is complementary to the prior identification of a high-fidelity Gly64Ser mutation in poliovirus 3Dpol that also resulted in reduced viral fitness in mice, and together there studies show that virus growth can be attenuated by using protein engineering to either increase or decrease viral polymerase fidelity.

In a recent new avenue for the lab, we developed a system for expressing and assaying the activity of the poliovirus 2B and 2C proteins using small lipid bilayers known as nanodiscs (2013). These proteins are bound to membranes and have been extremely difficult to express and purify for biochemical and structural studies using traditional bacterial or baculovirus based expression systems, but we found that coupled in vitro transcription and translation in the presence of nanodiscs yielded protein in sufficient quantities. In this work, we showed that the ATPase activity of the 2C protein is activated ≈20-fold upon cleavage from the 2BC precursor protein, leading to a model where 2C initially participates in viral replication complex formation via a low-activity and aggregation prone 2BC precursor whose ATPase activity is then activated upon cleavage to generate the fully processed 2C.

In addition to our core work on polymerase structure-function, we have also developed a set of fluorescence based assays for studying polymerase-RNA interactions and elongation. This began with a fluorescence polarization assay for simple RNA binding that was also sensitive to RNA elongation (2007), leading to the further  development of both a rapid stopped-flow kinetics based elongation assay (2009) and a high-throughput screening assay to identify polymerase inhibitors (2011). These assay have supported work looking at NTP stabilization of polymerase structure (2007) and the effects of mutations on elongation complex stability (2010) and RNA binding channels in the protein (2012).