Dynamic imaging of hepatitis C virus RNA localisation and traffic during viral replication

Abstract

Much of our understanding of the HCV life cycle and host-viral interactions has evolved from the visualisation of fixed images of infected cells. However, the recent development of live cell imaging techniques now allows viral life cycles to be visualised in live cell cultures. We have tagged the NS5A protein of the infectious Jc1 chimera (J6/JFH-1) with fluorescent tags and shown that NS5A segregates into two distinct populations: one relatively static and one highly motile, although the role and composition of these structures is not well understood. To investigate HCV RNA dynamics throughout the viral life cycle and examine whether either or both sub-classes of NS5A-positive structures are enriched with HCV RNA we developed a system to simultaneously track HCV RNA and NS5A in living cells. MS2 bacteriophage RNA stem loop sequences (6x /8x /12x /24x repeats) were inserted into the 3’UTR of the Jc1/5A-TCM virus (Jc1/5A-TCM+3’UTR:MS2) to allow indirect tracking of HCV RNA in Huh-7.5 cells via MS2.Coat-mCherry fusion protein that interacts specifically with MS2 stem loops. Jc1/5A-TCM+3’UTR:MS2 viruses replicated to significantly lower levels than the parent Jc1 as assessed by immunofluorescence analysis. However, long-term culture resulted in emergence of more efficient viral replication, with PCR and sequence analysis indicating at least partial retention of MS2 stem loops at 8 days post electroporation of HCV RNA. To further characterize and overcome the replication handicap induced by the insertion of the MS2 stem loop sequences we also generated Huh-7.5 cells that harbour the HCV subgenomic replicon featuring these MS2 stem loops insertions. Deep sequencing analysis was conducted to identify emerging adaptive mutations. However none was found to be particularly predominant. Most importantly, redistribution of the mCherry tagged-MS2 coat protein from a homogenous cytoplasmic distribution to a more punctate localisation was observed in the context of the full-length viral cultures indicating specific binding to HCV RNA. Using this approach we have simultaneously visualised HCV RNA (MS2.coatmCherry) and NS5A traffic (FlAsH) in real-time during HCV replication. Both HCV RNA-positive small motile and larger static structures were enriched with NS5A. In contrast, a subset of the trafficking NS5A-positive structures was devoid of HCV RNA. We also investigated viral RNA traffic with respect to lipid droplets (LDs) and show that two sub-types of static HCV RNA-positive structures existed: one was closely juxtaposed to LDs while the second sub-class was localised away from LDs. Moreover the system enabled visualization of putative RNA delivery at the LD surface with examples of motile HCV RNA-enriched structures dynamically interacting with LDs. Finally performing co-imaging of HCV NS5A and Rab18, an NS5A-interacting host factor located at the LD surface, we were able to illustrate the often transient nature of NS5A interaction with the LD and putative sampling of the LD that may precede interaction with core and initiation of assembly steps of the viral life cycle. These studies reveal new insights into the dynamics of HCV RNA traffic and the interactions at play in the context of the HCV life cycle.