A cooperative interaction between nontranslated RNA sequences and NS5A protein promotes in vivo fitness of a chimeric hepatitis C/GB virus B.

Abstract

GB virus B (GBV-B) is closely related to hepatitis C virus (HCV), infects small non-human primates, and is thus a valuable surrogate for studying HCV. Despite significant differences, the 5' nontranslated RNAs (NTRs) of these viruses fold into four similar structured domains (I-IV), with domains II-III-IV comprising the viral internal ribosomal entry site (IRES). We previously reported the in vivo rescue of a chimeric GBV-B (vGB/III(HC)) containing HCV sequence in domain III, an essential segment of the IRES. We show here that three mutations identified within the vGB/III(HC) genome (within the 3'NTR, upstream of the poly(U) tract, and NS5A coding sequence) are necessary and sufficient for production of this chimeric virus following intrahepatic inoculation of synthetic RNA in tamarins, and thus apparently compensate for the presence of HCV sequence in domain III. To assess the mechanism(s) underlying these compensatory mutations, and to determine whether 5'NTR subdomains participating in genome replication do so in a virus-specific fashion, we constructed and evaluated a series of chimeric subgenomic GBV-B replicons in which various 5'NTR subdomains were substituted with their HCV homologs. Domains I and II of the GBV-B 5'NTR could not be replaced with HCV sequence, indicating that they contain essential, virus-specific RNA replication elements. In contrast, domain III could be swapped with minimal loss of genome replication capacity in cell culture. The 3'NTR and NS5A mutations required for rescue of the related chimeric virus in vivo had no effect on replication of the subgenomic GBneoD/III(HC) RNA in vitro. The data suggest that in vivo fitness of the domain III chimeric virus is dependent on a cooperative interaction between the 5'NTR, 3'NTR and NS5A at a step in the viral life cycle subsequent to genome replication, most likely during particle assembly. Such a mechanism may be common to all hepaciviruses.