Partial nucleotide sequence of South American yellow fever virus strain 1899/81: structural proteins and NS1

We have partially cloned and sequenced the genome of a Peruvian yellow fever virus isolate (1899/81) and compared the nucleotide and deduced amino acid sequences of this strain with the previously published sequence of the West African yellow fever virus strain Asibi. In the 3594 base region sequenced, which contains the structural genes (C, M, E), all but the 72 3'-terminal nucleotides of the NS1 gene and 108 nucleotides of the 5' non-coding region, 515 nucleotide substitutions were detected. Nucleotide divergence was lowest in the 5' non-coding region, 2.8%, compared with an average rate of 14.7% in the coding regions. Over 91% of the 512 nucleotide changes in the coding region were silent; 44 amino acid substitutions resulted. The capsid protein was the least conserved, whereas the M protein was the most highly conserved (6.7% and 1.3% divergence, respectively). The envelope protein had 18 amino acid changes (3.7% divergence), one of which created an additional site for potential glycosylation of the 1899/81 virus. NS1 protein divergence (3.9%) was similar to that seen in the E protein. Of the 44 amino acid substitutions found, 34 (77%) were conservative. The highest number of nonconservative differences occurred in the envelope glycoprotein. These changes may significantly affect the antigenic and biological functions of the viruses.