Our research

Our laboratory studies the relationship between RNA viruses and their host cell, using hepatitis C virus (HCV) and dengue virus (DENV) as model systems.

All viruses are obligate parasites--that is, they are completely dependent on a host cell to complete their infection cycle.

We are generally interested in identifying cellular factors and/or pathways that support the viral infection cycle, and have used a number of different strategies to probe these relationships, including unbiased genetic and proteomic screens. Our work has help to uncover cellular pathways that are necessary to build viral replication organelles.

All positive-sense RNA viruses replicate their RNA genomes in association with an altered host membrane compartment. These viruses must reorganize a cellular membrane compartment to form specialized viral replication organelles. Inhibiting essential host cofactors of the viral infection cycle may represent a novel strategy for viral therapy. In principle, inhibiting host cofactors may present a higher genetic barrier for viral resistance than inhibiting viral proteins themselves.

1. Hepatitis C virus (HCV)

We have identified a number of host proteins that are required for HCV replication using a whole-genome RNAi screen (Tai et al, 2009). One of these proteins is a host phosphatidylinositol 4-kinase (PI 4-kinase) called PI4KA. PI 4-kinases regulate intracellular membrane trafficking through the local production of PI 4-phosphate (PI(4)P). PI4KA is recruited to HCV replication organelles, which then catalyzes the production of PI(4)P. PI4KA is required both for HCV replication as well as for integrity of the replication organelle. Several other RNA viruses also exploit host PI 4-kinases for their replication, indicating a common dependency for this pathway by multiple RNA viruses.

One major function of PI(4)P in cells is to serve as a signal for the recruitment of PI(4)P-binding proteins to membranes. Our laboratory has identified oxysterol-binding protein (OSBP) as an effector of PI4KA and PI(4)P in the establishment of HCV replication organelles (Wang et al, 2014). Interestingly, OSBP is a lipid transfer protein that exchanges cholesterol on one membrane for PI(4)P on another. The HCV replication organelle is enriched in cholesterol, and acute cholesterol depletion or inhibition of cholesterol biosynthesis inhibits HCV replication. Until this work, the mechanisms by which cholesterol is delivered to the RO were unknown. We have shown that OSBP is recruited to PI(4)P-enriched HCV replication organelles in a PI4KA-dependent manner, and that both PI4KA and OSBP are required for delivery of cholesterol to the HCV replication organelle. Therefore, our work has revealed a mechanism by which local generation of PI(4)P by PI4KA at viral replication organelles drives recruitment of OSBP and therefore cholesterol transport to sites of HCV replication.

More recently, we have studied the temporal dynamics of replication organelle formation (Wang and Tai, 2017). The question we sought to answer is whether replication organelles are continually renewed by newly-synthesized components, or whether they undergo a process of de novo formation, maturation, and turnover. To answer this question, we used optical pulse-chase methods to label newly synthesized NS5A and old NS5A molecules in living cells. We have found that HCV replication organelles in fact are formed de novo at spatially distinct sites from old replication organelles. This process is blocked by inhibitors of PI4KA, OSBP, and NS5A. As replication organelles mature, they appear to be preferential delivery sites for cholesterol and the ratio of positive:negative strand viral RNA increases. In addition, old replication organelles are preferentially found in association with lipid droplets and HCV core protein, and this association is blocked by cholesterol depletion.

We have also identified novel host factors that interact with the HCV nonstructural protein NS5A using a mass spectrometric approach, one of which is Rab18 (Salloum, Wang, et al, 2013). Rab18 associates with lipid droplets, which are cellular organelles for the storage of neutral lipids and cholesterol esters. However, it is also believed that HCV assembly occurs at or near lipid droplets. A significant question in the field is how progeny RNA genomes (presumably generated at replication organelles) become packaged into virions at lipid droplets. Our data suggest that the Rab18-NS5A interaction may help position sites of viral RNA replication near lipid droplets.

2. Dengue virus (DENV)

Dengue virus is the most common arboviral infection of humans, with estimates of nearly 400 million infections around the world each year. While it is largely a disease of the tropics and subtropics, the potential range for the two mosquito vectors for DENV is now estimated to cover much of the southern United States.

Our laboratory has completed a whole-genome CRISPR screen for host factors essential for DENV infection (Lin et al, 2017). This work confirms results of other similar screens in identifying the host oligosaccharyltransferase (OST) complex as a host dependency factor for DENV infection. The OST complex catalyzes the first step of protein N-glycosylation in the ER. There are two OST complexes bearing either the catalytic subunit STT3A or STT3B. Interestingly, ablation of either STT3A or STT3B blocks DENV infection but does not affect cellular viability or growth. We and others have found, however, that the N-glycosylation activity of the OST complex is dispensable for DENV infection. Instead, we found that the oxidoreductase activity of the MAGT1 subunit of the STT3B-containing OST complex is necessary for DENV infection. STT3B expression is essential for MAGT1 stability, and thus we hypothesize that the dependency of DENV infection on STT3B may be actually due to dependency on MAGT1.

We have also conducted a whole-genome transposon insertional mutagenesis screen to identify sites in the DENV genome that are tolerant of short 5 amino acid insertions (Perry et al, 2018). We hope that this will serve as a resource for the development of DENV genomes expressing functionally tagged viral proteins.

Our research has been made possible by the generous support of the National Institutes of Health, the Michigan GI Peptide Research Center, and the Greenview Foundation's Hepatitis C Research Fund.