Combining CRISPR-Cas9 base editing and proteomics to determine how filovirus VP24 proteins modulate ubiquitination and viral replication

Summary The filovirus family includes deadly, emerging/re-emerging, zoonotic viruses such as Marburg virus (MARV) and Ebola virus (EBOV). A recently described filovirus is Měnglà virus (MLAV) which was identified in a Rousettus bat in China. VP24 proteins are unique to the filoviruses, engage key cellular signaling pathways and play critical roles in viral infectivity and transmission. The VP24s from different filovirus lineages target different host pathways. EBOV VP24 binds importin alpha (IMPα) proteins that mediate STAT1 nuclear import, thereby blocking cellular responses to interferons (IFNs). MARV VP24 interacts with Keap1, a master regulator of antioxidant responses that recruits Cullin 3-Roc1 E3 ubiquitin (Ub) ligase to direct the degradation of transcription factor Nrf2. MARV VP24-Keap1 interaction frees Nrf2 from degradation, triggering a similar cytoprotective state. We found that MLAV VP24 binds neither IMPα nor Keap1. We hypothesized that MLAV VP24 instead targets distinct cellular signaling molecules. To address this, we performed comparative affinity purification-mass spectroscopy (AP-MS) proteomics in Rousetttus cells, examining the EBOV, MARV and MLAV VP24s. As expected, this identified IMPα and Keap1 as EBOV and MARV VP24 interactors, respectively. Consistent with our hypothesis, we identified the host deubiqutinating enzymes USP15, USP11 and USP4, which are paralogs, as unique MLAV VP24 interactors. Interestingly, USP15 regulates type I IFN responses and Nrf2-dependent antioxidant responses, among many other pathways. Because MARV and MLAV VP24s engage regulators of ubiquitination, we propose to define the Ub-regulated pathways targeted by each VP24, identify any overlap in the pathways affected and measure how targeting of ubiquitination affects MARV and MLAV infection. To this end, we will first perform mutational mapping to test the hypothesis that MLAV VP24 targets a region conserved between USP15, USP11 and USP4 and to identify loss of binding mutants. We will use inducible lentiviral vectors to deliver wildtype and loss of binding (to Keap1 or USPs) mutants for MARV and MLAV VP24 into Rousettus and human cells, determine whether these affect deubiquitination of known USP15, USP11 and USP4 targets and evaluate the impact of each VP24 on global ubiquitination in untreated, IFNα-treated and polyI:C transfected cells. We will then use CRISPR-Cas9 base editing to generate enzymatically inactive endogenous USP15, USP11, and USP4 and to disrupt Keap1 interaction with Nrf2. In parallel, we will use CRISPR-Cas9 knockout of the same host factors. We will use these cells in combination with proteomics profiling of Ub to define the global effects of these enzymes on host pathways. Finally, the base edited and knockout cells will be used, in combination with MARV and MLAV replication cycle modeling assays that enable study of filovirus replication at Biosafety Level 2, to determine the effect of these host enzymes on MARV and MLAV replication.