Exploiting the SARS-CoV-2 nsp14 3′-5′-exoribonuclease as a target for antiviral chemotherapy

RNA viruses exhibit high mutation rates due to the lack of proof-reading by their RNA-dependent RNA polymerases, imposing a restriction on genome size. In contrast to other positive-strand RNA viruses, coronaviruses have large genomes (~30kb). To ensure high fidelity replication the coronavirus non-structural protein 14 (nsp14) possesses 3'-5'-exoribonuclease (ExoN) activity, which also renders coronaviruses resistant to mutagenic nucleoside analogues (eg ribavirin). Nsp14 ExoN activity is stimulated by nsp10 binding, and structures of the SARS-CoV nsp14:nsp10 complex have been determined to 3.2Å resolution. Importantly, SARS-CoV and SARS-CoV-2 nsp14:nsp10 exhibit 95% amino-acid conservation, allowing us to build a homology model of SARS-CoV-2 nsp14:nsp10. In conjunction with protein dynamic simulations, this model will be subjected to virtual screening to identify small molecules able to bind to either the ExoN active site, or the nsp14:nsp10 interface. Compounds will be tested in combination with mutagenic nucleoside analogues for effects on the replication of SARS-CoV-2, exploiting in-house BSL-3 containment facilities. Specificity and mode of action of compounds will be confirmed by virological and biochemical assays. Our priority is to seek existing clinically approved compounds that inhibit the SARS-CoV-2 ExoN, and could be rapidly repurposed to treat SARS-CoV-2 infection. To this end, we will interrogate the DrugBank database of approved or investigational drugs. This will be followed by an expansion of virtual screening to an in-house library, and a ZINC library subset, comprising available drug-like compounds. In the longer term these could lead to early-stage drug discovery to provide further therapeutic options for SARS-CoV-2, or other future emerging coronaviruses.