Secondary structure determination of the conserved 98-base sequence at the 3' terminus of hepatitis C virus genome RNA

Transcripts of a chimeric cDNA clone of hepatitis C virus genotype 1b are infectious in vivo

We constructed a chimeric cDNA clone of hepatitis C virus (HCV) that is infectious. The chimeric genome encodes the polyprotein of a genotype 1b strain (HC-J4) of HCV and replicates via 5' and 3' untranslated regions of a genotype 1a strain. The infectivity of three full-length cDNA clones was tested by direct injection of RNA transcripts into the liver of a chimpanzee. The chimpanzee became infected with HCV and the viral titer increased over time from 10(2) genome equivalents (GE)/ml at week 1 postinoculation (p.i.) to 10(4)-10(5) GE/ml during weeks 3-11 p.i. Antibodies to HCV were detected from week 18 p.i. However, the chimpanzee did not develop hepatitis. Sequence analysis of PCR products amplified from the serum of the chimpanzee demonstrated that only one of the three clones was infectious. Sequence comparisons with the cloning source, an acute-phase infectious plasma pool derived from an experimentally infected chimpanzee, showed that this infectious clone had three amino acids that differed from the consensus sequence of HC-J4, whereas the two noninfectious clones had seven and nine amino acid differences, respectively. Together, genotype 1b, represented by the infectious molecular clone described herein, and genotype 1a, represented by the two cDNA clones previously shown to be infectious for chimpanzees, account for the majority of HCV infections in the United States, Europe, and Japan.

Ubiquitous presence of cellular proteins that specifically bind to the 3' terminal region of hepatitis C virus

The 3' terminal region (3'-X tail) of hepatitis C virus (HCV) genomic RNA forms a stable stem-loop structure. The 3'-X tail consists of 98 nucleotides (nt) that are highly conserved among the HCV strains and supposed to function as a cis-acting region for replication of negative strand RNA and/or viral encapsidation. In the present study, by UV cross-linking assay we found two kinds of cellular proteins of approximately 87 and 130 kDa, which specifically bind to the full-length 3'-X tail (nt 1 to 98), but not the 3'- or 5'-truncated 3'-X tail, consisting of nt 1 to 50 or nt 51 to 98, respectively. These proteins were detected in human cell lines such as hepatic tumor cell lines and a T-lymphocyte cell line and also in a human embryonic lung fibroblast cell strain. In addition, human hepatocellular carcinoma tissues expressed these proteins regardless of infection or uninfection of HCV. Furthermore, these proteins were also detected in normal human tissues derived from the lung, heart, kidney, stomach, intestine, and colon. Thus, these cellular proteins, which are ubiquitously present in human tissues, might be involved in viral replication and/or encapsidation.