Nucleotide sequences of hepatitis GB virus C. Identification of highly conserved domains in the 5' noncoding region and detection by polymerase chain reaction

We detected the hepatitis GB virus C genome by reverse trnascription-polymerase chain reaction in three Japanese patients with chronic liver diseases. Partial nucleotide sequences of 5' noncoding, envelope and NS3 regions had 88.6-90.1: 86.388.0, and 78.6-79.6% nucleotide sequence homology compared with the prototype GBV-C genome. However, they showed higher homology with each other (96.197.1, 88.7-91.2, 84.0-87.0%, respectively), suggesting that they were a genotype of GBV-C. The domains that were highly conserved among all these genomes were present in the 5' noncoding region. The frequency of detecting the genome by the polymerase chain reaction was higher when we used primers designed on these domains compared with primers designed on the core and NS3 region. The virus genome was detected in eight of 50 (16.0%) consecutive Japanese patients with hepatocellular carcinoma who had received more than ten units of blood transfusion, three of 60 (5.0%) patients with non-B, non-C liver disease and 12 of 35 (34.3%) Malaysian patients with non-B chronic liver disease. Since there are no reliable assays to detect hepatitis GB virus C at present, the detection of the genome by the polymerase chain reaction should be useful for diagnosis. Further nucleotide sequence analysis of the genomes is necessary for epidemiological survey studies and vaccine strategy. (C) 1997 Elsevier Science Ireland Ltd.