Binary selectivity for HCV NS3/4A protease versus host proteases via irreversible inactivation at a non‐catalytic cysteine

Abstract

Viral protease inhibitors typically achieve potency via interactions with the catalytic apparatus, but as a result are often poorly selective. Viral protease inhibitors must also block emerging mutants. We have developed a fundamentally different approach to the design of covalent inhibitors by targeting non‐catalytic cysteine residues unique to viral targets but absent in host proteases, enabling extreme, “binary” selectivity. An estimated 170 million people worldwide are infected with hepatitis C virus (HCV). HCV NS3/4A protease (HCVP) is a key target for HCV therapy. We have identified a class of irreversible HCVP inhibitors that bond covalently to the non‐catalytic Cys159 in HCVP. Cys159 is conserved across all HCV genotypes but is absent in host proteases. MALDI‐TOF MS and in vitro assays show that the lead inhibitor bonds to and inhibits HCVP across multiple genotypes and drug‐resistant variants. In replicon assays covalent silencing leads to potent inhibition of HCVP and prolonged duration of action (>24 hrs) after brief exposure to inhibitor. Covalent silencing yields potent inhibition of HCVP in vivo, in a new mouse model. Inhibitors with this profile may profoundly improve hepatitis C patient treatment across all HCV genotypes. This design strategy is also readily applicable to a wide range of therapeutically important proteases. This work was supported by Avila Therapeutics.