Proof of principle for a CRISPRi screen to identify the host protease cleaving the NS1/2A site in the yellow fever virus (YFV) polyprotein

Abstract

Mosquito-borne flaviviruses like yellow fever (YFV), dengue (DENVs), and Zika viruses (ZIKV) affect close to half a billion people every year and contribute significantly to human morbidity and mortality worldwide. There are no approved directly acting antiviral therapies, and preventive vaccines are only available for few flaviviruses. The positive-strand RNA genome of flaviviruses encodes a single polyprotein, which is proteolytically processed into 10 individual proteins by viral and host proteases. While most proteases catalyzing cleavage events in the flavivirus polyprotein have been identified, the protease cleaving the junction between two non-structural proteins, NS1 and NS2A, is still unknown. Cleavage at this site is essential for YFV viral replication, making the protease—known to be an ER-associated host protease—a potential drug target. A screen using Clustered Regularly Interspaced Short Palindromic Repeats interference (CRISPRi) technology to downregulate all cellular proteases and identify cells that do not support YFV infection via fluorescence-activated cell sorting (FACS) could identify candidate proteases involved in the NS1/2A cleavage. In this thesis, we explore the feasibility of such a screen and establish multiple proof of principles in support of it. We created a YFV-17D genome variant with mutations in the highly conserved octapeptide motif at the junction of NS1 and NS2A, which contains a recognition sequence critical for proteolytic cleavage. We found that mutations in this motif reduced viral replication kinetics in human hepatoma (Huh7) cells akin to levels of a genome harboring mutations in the active site of the viral RNA dependent RNA polymerase. The fact that the octapeptide motif in the sequence directly preceding the NS1/2A cleavage site is conserved among flaviviruses suggests that the protease cleaving YFV NS1/2A might also be responsible for this site’s cleavage in other flaviviruses, making it an interesting target for antivirals with broad-spectrum activity. We also use an optimized CRISPRi experimental approach to knockdown subunits of the signal peptidase complex (SP), a protease involved in YFV polyprotein processing. We noticed a detectable difference in the mean fluorescent intensity (MFI) of a reporter mScarlet-YFV-17D virus when the SPCS1 subunit of SP is knocked down. Further optimization is required before the CRISPRi screen targeting all human proteases can be performed. Collectively, this work provides a foundation for further refinement of genetic loss-of-function approaches aimed at identifying the elusive cellular protease responsible for NS1/NS2 cleavage.