IDENTIFICATION OF GB VIRUS-C VARIANTS BY PHYLOGENETIC ANALYSIS OF 5'-UNTRANSLATED AND CODING REGION SEQUENCES

Phylogenetic analysis of 44 GB virus C (GBV-C) 5'-untranslated region (5'-UTR) sequences from 37 individuals suggested the presence of GBV-C genotypes (A. S. Muerhoff, J. N. Simons, T. P. Leary, J. C. Erker, M. L. Chalmers, T. J. Pilot-Matias, G. J. Dawson, S. M. Desai, and I. K. Mushahwar, J. Hepatol. 25:379-384, 1996) that correlated with geograp hic origin: type 1, 2a and 2b, and 3 isolates are found predominantly in West Africa, the United States and Europe, and Japan, respectively. We have extended our analysis to include 5'-UTR sequences from 129 gl obally distributed GBV-C isolates and sequences from the second envelo pe protein (E2) gene and nonstructural (NS) regions 3 and 5b from a su bset of these isolates. Bootstrap analysis of a 157-nucleotide segment of the 5'-UTR from 129 sequences provided weak support for the existe nce of the four major groups of GBV-C isolates previously described, a lthough phylogenetic analysis of a 374-nucleotide segment of the 5'-UT R from 83 isolates provided stronger support. Thus, the groups of GBV- C variants previously identified upon analysis of the entire 5'-UTR ca n be distinguished by analysis of the shorter, 374-nucleotide region f rom the 5'-UTR. In contrast, independent analysis of the E2, NS3, or N S5b region sequences does not identify groups of GBV-C variants that c orrelate with geographic origin. However, bootstrap analysis of these coding sequences, when linked to form colinear sequences, demonstrates that longer coding regions can produce GBV-C groupings that are simil ar to that determined from 5'-UTR sequence analysis. The inability to distinguish between GBV-C variants by using small segments of coding s equence suggests that the GBV C genome is constrained. As a result of these constraints, there is a high degree of nucleotide and amino acid sequence conservation between isolates from widely separated geograph ic areas. Hence, substitutions at many nucleotide positions are not to lerated, so that substitutions at the positions which tan change are s aturated, thereby obscuring the evolutionary relationships.