Using viral replicative infidelity to test viral proteases

Two of the most important characteristics of positive-strand RNA viruses are the error-prone nature of their polymerases and their dependence on proteinases to specifically cleave larger precursors into the mature viral proteins. This project sets out to utilise the error-prone nature of RNA viral replication to generate substantial new information on the properties of proteinases encoded by human immunodeficiency virus (HIV) and hepatitis C virus (HCV), two viral pathogens of global importance. In spite of the success of anti-retroviral therapy to treat AIDS patients, many open questions remain. These include understanding how the HIV proteinase becomes resistant to inhibitors and the nature of the substrate specificity of the enzyme itself. Similar questions affect possible therapies targeting the NS3/4 proteinase of HCV. Control of HCV infections by proteinase inhibitors is in its infancy, with one inhibitor only recently entering clinical trials. Further information on the substrate specificity and the question of resistance would be very important for development of substances inhibiting HCV replication. To attempt to answer these questions, we propose to establish a novel assay system for viral proteinases using an RNA replicon from the flavivirus tick-borne encephalitis virus (TBEV). RNA virus replicons are RNA molecules encoding the viral proteins and RNA signals for viral replication, but lacking one or more of the structural proteins. The RNA molecules replicate, but do not produce virus-like particles (VLPs) unless the missing structural proteins are provided in trans by the cell in which the replicon is replicating. In this project, the replicons will be designed so that infectious particle formation, which can be sensitively measured in cell culture, depends on a specific cleavage by the HIV or HCV protease NS3/4. Specifically, the viral proteinases will be expressed from the replicon. Sequences corresponding to the viral cleavage sites will be introduced between the capsid protein C and the surface protein prM and the modified polyprotein precursor to these structural proteins will be stably expressed in BHK cells. Only when the TBEV structural proteins are cleaved by the HIV or HCV proteinase will VLPs be generated. The inherent error-rate of viral RNA polymerases will ensure that the sequences encoding the viral proteinases will also be mutated. If the cleavage site of the proteinase is sub-optimal or a proteinase inhibitor is present, VLP production will be low. Proteinase mutants generated by the viral polymerase should have a selective advantage and be able to cleave the sub-optimal cleavage site or be resistant to the inhibitor. This will lead to increased VLP production. Analysis of the VLPs will allow the mutations responsible to be identified, thus generating information on which amino acids of the proteinase interact with substrates and inhibitors. The replicon system will thus be a sensitive assay for the activity and inhibition of the HIV and HCV proteinases, enabling rapid screening of the substrate specificity as well as the rate of appearance of proteinases which are resistant to an inhibitor.

Two of the most important characteristics of positive-strand RNA viruses are the error-prone nature of their polymerases and their dependence on proteinases to specifically cleave larger precursors into the mature viral proteins. This project set out to utilise the error-prone nature of RNA viral replication to establish a novel assay system for viral proteinases. We used an RNA replicon from the tick-borne encephalitis virus (TBEV). RNA virus replicons are RNA molecules encoding the viral proteins and RNA signals for viral replication, but lacking one or more of the structural proteins. These RNA molecules replicate, but do not produce virus-like particles (VLPs) unless the missing structural proteins are provided in trans by the cell in which the replicon is replicating. In this project, the replicons were designed so that infectious particle formation, which can be sensitively measured in cell culture, depends on a specific cleavage by the foot-and-mouth disease virus (FMDV) 3C proteinase. To establish this system, we first examined the flexibility of a region of TBEV between the protein C and the protein prM. We found that the virus can unexpectedly tolerate a large number of substitutions at this position and that even substantial deletions could become viable by the selection of natural variants. We also ascertained that cleavage between C and prM was possible through the introduction of a sequence of 18 amino acids of the 2A protein from the FMDV virus. This sequence is known to interrupt a growing polypeptide chain, although the mechanism of interruption remains unclear. Viruses containing this sequence were viable and capable of infecting other cells. Thus, it is possible to decouple cleavage of the C-prM junction from the normal NS2B/3 proteinase cleavage. We also extended this work to show that a similar result could be observed with the related West Nile virus (WNV). We also investigated the properties of these viruses containing the 2A sequence of FMDV. The TBEV mutant had a defect in entering mammalian cells and was incapable of replication in the tick cell. In contrast, the WNV mutant could infect the natural host mosquito cells but was unstable in mammalian ones. Removal of three specific amino acids from the 2A protein sequence in both the TBEV and WNV mutant viruses led to replication of the viral RNA; however, viable virus was not produced. To overcome this block, we expressed the FMDV 3C proteinase from a second coding region at the 3` end of the viral genome. When the N-terminal sequence of the FMDV 2A was optimized for FMDV 3C proteinase specificity, we found that viable virus particles were produced. In contrast, when an inactive proteinase mutant was used, no particles could be detected. This therefore shows that the viral replication is now dependent on the activity of a heterologous proteinase. In the future, this system will be extended to other viral proteinases such as those of poliovirus and HIV. We also attempted to determine the crystal structure of the TBEV NS2B/3 proteinase. However, despite our success in the expression in bacteria of milligram quantities of pure NS2B/3 protein, we were unable to obtain reproducible crystals.