Sequence Heterogeneity within Three Different Regions of the Hepatitis G Virus Genome

AbstractTwo sets of primers derived from the 5′-terminal region and the NS5 region of the hepatitis G virus (HGV) genome were used to amplify PCR fragments from serum specimens obtained from different parts of the world. All PCR fragments from the 5′-terminal region (5′-PCR,n= 56) and from the NS5 region (NS5-PCR,n= 85) were sequenced and compared to corresponding published HGV sequences. The range of nucleotide sequence similarity varied from 74 and 78% to 100% for 5′-PCR and NS5-PCR fragments, respectively. Additionally, five overlapping PCR fragments comprising an approximately 2.0-kb structural region of the HGV genome were sequenced from each of five sera obtained from three United States residents. These sequences were compared to 20 published sequences comprising the same region of the HGV genome. Nucleotide and deduced amino acid sequences obtained from different individuals were homologous from 82.9 to 93.6% and from 90.4 to 99.0%, respectively. Sequences obtained from follow-up specimens were almost identical. Comparative analysis of deduced amino acid sequences of the HGV structural proteins and hepatitis C virus (HCV) structural proteins combined with an analysis of predicted secondary structures and hydrophobic profiles allowed prediction of processing sites within the HGV structural proteins. A phylogenetic sequence analysis performed on the 2.0-kb structural region supports the existence of three previously identified HGV genetic groups. However, phylogenetic analysis performed on only small DNA fragments yielded inconsistent genetic grouping and failed to confirm the existence of genetic groups. Thus, in contrast to HCV where almost any region can be used for genotyping, only large or carefully selected genome fragments can be used to identify consistent HGV genetic groups