Two viruses, GB virus A (GBV-A) and GB virus B (GBV-B), were recently identified in the GB hepatitis agent. Human sera containing antibodies that recognize GBV-A and/or GBV-B recombinant proteins were subjected to polymerase chain reaction studies with degenerate oligonucleotides capable of amplifying a segment of the putative helicase genes from GBV-A, GBV-B or hepatitis C virus. Novel sequences related to members of the Flaviviridae were identified in sera from 12 individuals including 4 individuals with hepatitis. The limited nucleotide sequence identity between GBV-A, GBV-B and HCV sequences suggests that a novel virus, tentatively named GB virus C, may be responsible for some cases of non-A, non-B, non-C, non-D, non-E hepatitis.

A subtractive PCR methodology known as representational difference analysis was used to clone specific nucleotide sequences present in the infectious plasma from a tamarin infected with the GB hepatitis agent. Eleven unique clones were identified, seven of which were examined extensively. All seven clones appeared to be derived from sequences exogenous to the genomes of humans, tamarins, Saccharomyces cerevisiae, and Escherichia coli. In addition, sequences from these clones were not detected in plasma or liver tissue of tamarins prior to their inoculation with the GB agent. These sequences were detected by reverse transcription-PCR in acute-phase plasma of tamarins inoculated with the GB agent. Probes derived from two of the seven clones detected an RNA species of .8.3 kb in the liver of a GB-agent-infected tamarin by Northern blot hybridization. Sequence analysis indicated that five of the seven clones encode polypeptides that possess limited amino acid identity with the nonstructural proteins of hepatitis C virus. Extension of the sequences found in the seven clones revealed that plasma from an infected tamarin contained two RNA molecules >9 kb long. Limited sequence identity with various isolates of hepatitis C virus and the relative positions of putative RNA helicases and RNAdependent RNA polymerases in the predicted protein products of these molecules suggested that the GB agent contains two unique flavivirus-like genomes.

No abstract

Among the three recently described GB viruses (GBV-A, GBV-B, and GBV-C), only GBV-C has been linked to cryptogenic hepatitis in man. Because of the limited utility of currently available research tests to determine antibody response to GBV-C proteins, the prevalence of GBV-C RNA in human sera was studied using reverse transcription-polymerase chain reaction (RT-PCR). The prevalence of GBV-C is higher among volunteer blood donors with elevated serum alanine aminotransferase (ALT) levels (3.9%) than among volunteer blood donors with normal ALT levels (0.8%). Higher rates were also noted among commercial blood donors (12.9%) and intravenous drug users (16.0%). GBV-C was frequently detected in residents of West Africa, where the prevalence was > 10% in most age groups. Approximately 20% of patients diagnosed with either acute or chronic hepatitis C virus (HCV) were found to be positive for GBV-C RNA. In addition, GBV-C RNA sequences were detected in individuals diagnosed with non-A-E hepatitis, with clinical courses ranging from mild disease to fulminant hepatitis. Fourteen of sixteen subjects with or without clinically apparent hepatitis were positive for GBV-C RNA more than 1 year after the initial positive result.

Two flavivirus-like genomes have recently been cloned from infectious tamarin (Saguinus labiatus) serum, derived from the human viral hepatitis GB strain, which is known to induce hepatitis in tamarins. In order to study the natural history of GB infections, further transmission studies were carried out in tamarins. Reverse-transcription-polymerase chain reaction and enzyme-linked immunosorbant assays were developed for the detection of RNA and antibodies associated with the two agents, GB virus-A and GB virus-B. The infectivity of both of these agents was demonstrated in tamarins to be filterable through a 0.1 micron filter. Two distinct genomes were identified in the serum of eight of the infected tamarins, while in four tamarins, the genomes were detected independently of each other. Although specific antibodies to the GB virus-B epitopes were detected in the serum of animals inoculated with both agents or GB virus-B alone, antibodies to putative epitopes specific to GB virus-A were not detected in any of the animals. All tamarins inoculated with serum containing GB virus-B exhibited an elevation in liver enzyme levels after inoculation. Elevations of serum liver enzyme levels did not occur when GB virus-A was the only agent detected in the serum. Infection with the original infectious tamarin inoculum conferred protection from reinfection with GB virus-B but not with GB virus-A.

The genomes of two positive-strand RNA viruses have recently been cloned from the serum of a GB agent-infected tamarin by using representational difference analysis. The two agents, GB viruses A and B (GBV-A and GBV-B, respectively), have genomes of 9,493 and 9,143 nucleotides, respectively, and single large open reading frames that encode potential polyprotein precursors of 2,972 and 2,864 amino acids, respectively. The genomes of these agents are organized much like those of other pestiviruses and flaviviruses, with genes predicted to encode structural and nonstructural proteins located at the 5 and 3 ends, respectively. Amino acid sequence alignments and subsequent phylogenetic analysis of the RNA-dependent RNA polymerases (RdRps) of GBV-A and GBV-B show that they possess conserved sequence motifs associated with supergroup II RNA polymerases of positive-strand RNA viruses. On the basis of similar analyses, the GBV-A-and GBV-B-encoded helicases show significant identity with the supergroup II helicases of positive-strand RNA viruses. Within the supergroup II RNA polymerases and helicases, GBV-A and GBV-B are most closely related to the hepatitis C virus group. Across their entire open reading frames, the GB agents exhibit 27% amino sequence identity to each other, approximately 28% identity to hepatitis C virus type 1, and approximately 20% identity to either bovine viral diarrhea virus or yellow fever virus. The degree of sequence divergence between GBV-A and GBV-B and otherFlaviviridae members demonstrates that the GB agents are representatives of two new genera within the Flaviviridae family.

Two overlapping sets of TT virus (TTV)-specific polymerase chain reaction primers were used to test for presence of TTV, which was found in approximately 10% of US volunteer blood donors, 13% of commercial blood donors, and 17% of intravenous drug abusers. The rate of TTV infection among US non-A, non-B, non-C, non-D, non-E hepatitis patients was only 2%. Among commercial blood donors and intravenous drug abusers, only 1%-3% of the TTV-positive individuals were coinfected with GB virus C (GBV-C), a parenterally transmitted virus. This suggests that GBV-C and TTV may have different routes of transmission. Comparison of the sensitivities of 2 TTV polymerase chain reaction (PCR) primer sets showed that the majority of samples were detected with only 1 of the 2 sets. Therefore, previous studies in which only a single PCR primer pair was used may have significantly underestimated the true prevalence of TTV.

A 336-amino-acid segment of the GB virus C second envelope protein (E2) has been produced in BHK-21 cells using the Semliki Forest virus vector system. Secretion of this protein was facilitated by deletion of a hydrophobic region at the C-terminus that may represent the membrane anchoring domain. The E2 protein recovered from the culture supernatant exhibited a molecular mass of approximately 52 kDa, with the increase in size relative to the polyprotein backbone being contributed by N-linked glycosylation. A radioimmunoprecipitation assay using GBV-C E2 was developed to test for the presence of antibodies against this protein in human sera. The prevalence of antibodies to E2 was high among injection drug users and other individuals at risk for acquiring parenterally transmitted agents. There was a much higher percentage of anti-E2 seropositivity in GBV-C RT-PCR negative compared to GBV-C RT-PCR positive samples from these populations. In addition, serial samples from patients transfused with blood containing GBV-C showed seroconversion to anti-E2 positivity and loss of GBV-C viremia as measured by RT-PCR within 11 months of transfusion in five of seven individuals. Thus, this system provided a rapid means to identify GBV-C E2 as a useful antigen for the study of GBV-C exposure.