Cause and effect: Hepatitis C virus infection and host innate immune response

The hepatitis C virus (HCV) infects as many as 180 million people worldwide and has recently been linked to more deaths in the United States than the human immunodeficiency virus (HIV). HCV is a positive-sense, single-stranded RNA virus of the family Flaviviridae and genus Hepacivirus. The first aim of my thesis was to characterize the nonstructural HCV protein NS3-4A. NS3-4A is comprised of NS3, a bifunctional protease/helicase, and NS4A, a single-spanning transmembrane (TM) protein that serves as a cofactor for the NS3 protease. The NS4A TM domain contains several invariant glycine and alanine amino acid residues commonly seen in other TM-TM interaction motifs, suggesting that the NS4A TM may be involved in specific homo- or hetero-typic interactions. Through biochemical and cell culture studies, I uncovered a specific homotypic interaction of the NS4A TM domain and demonstrated that this interaction is required for HCV replication and viral assembly. Furthermore, computational modeling of this interaction predicted that the NS4A TM domain forms a right-handed parallel homodimer (Chapter 2).;An important implication of an NS4A TM homodimer is the oligomerization of the NS3-4A protease/helicase. Modeling of the NS3-4A homodimer revealed that the conformation of the NS3 protease domain relative to the helicase domain is critical for dimer formation atop the ER membrane. Interestingly, earlier biochemical studies in the Pyle lab by Ding et al. showed that NS3 can adopt two conformations in vitro and that these two conformations have different functional attributes. To extend these findings in a more virological context, I used mutagenesis to systematically alter the flexibility and length of the linker region connecting the two domains of NS3 and assessed the effects in vitro and in cell culture. I found that these mutations interfered with NS3 toggling between conformational states, thus disrupting HCV replication and viral assembly (Chapter 3).;RIG-I, or retinoic acid-inducible gene 1, is a duplex RNA-activated ATPase (DRA) that plays a major role in the innate immune system of higher-order metazoans by recognizing pathogenic viral genomic RNA including HCV RNA. Fascinatingly, both RIG-I and HCV NS3 are superfamily 2 helicases with a similar overall domain arrangement of 2 RecA-like alpha-beta domains and an overlapping set of ATP and RNA binding motifs (I-VI). Despite having the same overall fold, RIG-I functions at the end of 5'triphosphorylated duplex RNA substrates without unwinding, whereas NS3 translocates on ssRNA substrates and unwinds duplex RNA regions. The accessory domains of RIG-I -- a HEL2i insertion and a C-terminal domain (CTD) -- and NS3 -- a C-terminal all-alpha domain -- are primarily responsible for the different biological functions of these two proteins, in addition to accessory motifs like motif 2a/motif Vc (top strand recognition) or the NS3 beta-hairpin motif (for separation of duplex RNA).;The second aim of my thesis was to define the minimal, functional PAMP (dsRNA region) required for RIG-I recognition and activation for both RIG-I ATPase activity and RIG-I-mediated interferon-beta production, and to determine the oligomerization state of RIG-I before and after RNA binding. Through structural, biophysical, biochemical, and cell culture studies, I demonstrated that RIG-I surveys the cytoplasm of cells as a monomer and binds to the 5' ends of duplex RNA as a monomer (Chapter 4). RIG-I requires only 10 base pairs at the end of duplex RNA for robust ATPase activation and interferon-beta stimulation. Interestingly, I found that RIG-I binding to the internal stem regions of duplex RNA is neither favored nor required, thereby explaining why RIG-I prefers shorter dsRNA compared to its paralog MDA5, which prefers longer dsRNA and forms cooperative filaments on internal stretches of dsRNA.