Mosquito-transmitted flaviviruses include important human pathogens such as Dengue virus (DENV), Yellow Fever virus (YFV), West Nile virus (WNV) and Japanese Encephalitis virus (JEV). These viruses contribute to millions of human cases of disease with significant mortality across the globe. Unfortunately, there are currently no effective drugs available for the specific treatment of flaviviral diseases. Disrupting critical interactions between flaviviral proteins involved in virus replication by peptidomimetics is one approach to developing effective antiviral therapies.
The non-structural protein 5 (NS5) is a large protein that is structurally conserved among flaviviruses, making it an attractive target for antiviral drug studies. The protein consists of the methyltransferase (MTase) and the RNA dependent RNA polymerase (POL) domains. Stability of the viral RNA is maintained by MTase via cap methylation of the 5’ end of nascent RNA while POL is pivotal for RNA synthesis. Amino acid mutations of residues 46, 47 and 49 in the MTase αA3-motif resulted in the attenuation of DENV. A compensatory mutation at residue 512 (Leu512 loop) in POL, which rescued viral replication of this mutant, was found after several passages of the virus, suggesting that the residues selected for mutation were crucial for replication. In addition, structural modelling of the putative full-length NS5 crystal structure showed a close proximity of the αA3-motif in the MTase to the Leu512 loop in POL, which suggested a physical interaction between the two domains via the αA3-motif. Taken together, these findings indicated a functional interaction between the MTase αA3-motif and the POL domain in virus replication. However, evidence for a physical interaction between these domains had not been documented. This study aimed to demonstrate a physical interaction between the MTase and POL domains of the WNV NS5 protein using recombinant MTase proteins as well αA3-like mimetic peptides and assessing their interactions with recombinant POL in vitro and in vivo (in cell). The potential of the mimetic peptides to inhibit a putative functional interaction between MTase and POL was also to be investigated.
To identify the potential interaction between the domains, a series of constrained and cyclic peptides were designed from the amino acid sequence of αA3-motif and their binding profiles to mAb 5H1 (a monoclonal antibody specific to the αA3-motif) and recombinant POL were assessed. While the peptides were recognized by mAb 5H1 in both ELISA and fluorescent polarization assay (FPA), indicating that they retained the native antigenic structure of the αA3-motif when bound to the solid phase or in solution, they failed to bind recombinant POL in ELISA and FPA. This led to the investigation into the interaction between full-length domains of NS5 expressed individually in human embryonic cells which more accurately represent the native environment for these recombinant proteins. Using pulldown assays, it was shown that the recombinant MTase protein specifically interacted with recombinant POL when co-expressed in the cytoplasm of the mammalian cell. This is the first evidence of a physical association between the MTase and POL domains of the flavivirus NS5 protein. In addition, full-length native NS5 co-expressed via a replicon in the cell was also pulled down by recombinant POL, suggesting the interaction was not an artifact of the recombinant MTase.
The next logical step was to assess the importance of residues in the αA3-motif as putative sites of interaction with recombinant POL in the context of the full-length recombinant MTase protein. Selected residues in the αA3-motif were mutated in the recombinant MTase and the effect on POL interaction was assessed by pulldown assays. These mutations were also introduced into the WNVKUN infectious clone (FLSDX) and the replication properties of these mutant viruses assessed. However the effects of these mutations on MTase-POL association were shown to be insignificant by pulldown assay, suggesting that these residues were not critical to the interaction. Similarly, all but one of these mutations had no effects on virus replication when introduced into the infectious clone. The exception (E46L mutation), completely abolished virus replication suggesting some requirement of this residue in replication. However failure to generate compensatory mutations in POL that rescued replication, even after serial passage of the transfection supernatant in Vero cells, precluded further conclusion of the significance of this result in the context of MTase-POL interactions.
To determine if the mimetic peptides could be used as potential inhibitors to the interaction between MTase and POL domains, attempts were made to develop a novel coupled MTase and POL assay using WNVKUN infected extracts. However, due to technical difficulties, this approach was discontinued. Instead an in vitro POL assay was established to assess whether the αA3-motif peptides could inhibit RNA synthesis by binding to POL. While a POL assay was successfully established in the lab, inhibition results using peptides and anti-POL antibodies were not reproducible, suggesting this was not an optimum system for assessing POL inhibitors.